Hi, everyone. A quick note before we start. This conversation was recorded before Viktor Orbán lost his reelection bid for Prime Minister of Hungary. After 16 years in power, he was defeated by Péter Magyar. Okay, now let’s start the show.

Kim Lane Scheppele:

The dominant way that democracies fail is through aspirational autocrats being elected in free and fair elections, coming to power, changing the law, usually in illegal fashion, until the next thing you know, there’s no way to get rid of the autocrat.

Katy Glenn Bass:

Hello and welcome to Lawyering Without Law, a new podcast by the Knight First Amendment Institute where we’ll explore the unique and important role that lawyers and the legal profession play in defending democracy or facilitating a country slide into authoritarianism. I’m Katy Glenn Bass, the research director at the Knight First Amendment Institute, and you’ve just heard a clip of a conversation with our first guest, Kim Lane Scheppele, who we’ll introduce in just a minute.

Joining me in co-hosting this podcast is Professor Madhav Khosla, the B. R. Ambedkar Professor of Indian Constitutional Law and Professor of Political Science at Columbia University and the Knight Institute’s senior fellow. Madhav, I’m really looking forward to hosting these conversations with you.

Madhav Khosla:

Thank you, Katy. I’m really looking forward to it as well. Over the next six episodes, we’ll bring both historical examples, comparisons and contrast to what we’re seeing now, as well as more specific conversations about democratic backsliding, both in the US and around the world, and in particular, the legal profession’s role in these developments over the last few years.

Joining us today on our very first episode to explore the use and deployment of the law and legal systems as a central feature of contemporary authoritarian populism is Kim Lane Scheppele, the Laurance S. Rockefeller Professor of Sociology and International Affairs at Princeton University. Professor Scheppele’s primary field is the sociology of law, and a lot of her research examines the rise and fall of constitutional governments.

She’s worked extensively on the undermining of civil liberties during the post 9/11 period, and in recent times has been integral to debates on the transition from democracy to authoritarianism. Professor Scheppele, welcome to Lawyering Without Law, and thank you for joining us.

Kim Lane Scheppele:

It’s wonderful to be here, although the thought of being without law is vaguely terrifying.

Madhav Khosla:

It is, and hopefully we can figure out exactly how that happens. So just to start us off, Kim, I was wondering if you could just share a little bit about your contemporary work on illiberal government and on democratic backsliding.

Kim Lane Scheppele:

Well, when I started working on this line of research, I actually didn’t think it was going to be about backsliding democracies. I thought I was looking at, well, I don’t know, frontsliding democracies, countries that were coming out from under Soviet tutelage that were setting up democracies first in what was Eastern Europe, then we called it Central Europe, and then in the former Soviet Union itself.

I worked for four years at the Hungarian Constitutional Court in the 1990s. I worked for one year at the Russian Constitutional Court in the early 2000s, and I thought what I was watching was the onset of democratic institutionalism, rule of law, and respect for rights. Now, oh, I guess I shouldn’t have been massively surprised. Both Viktor Orbán and Vladimir Putin, once they came to power, were lawyers and they understood how law worked.

And so as a result, when they decided they wanted to stay in power for the long term and not allow power to rotate, they wound up dismantling the very institutions that had just been set up and they did so by law. Because there were lawyers, they could work out what are the soft spots in the system, that if you pushed on them would generate dictatorship rather than democracy.

They worked with their parliaments to push laws through. They worked with courts to increase the number of judges actually, often on courts, packing them with their own people, and using perfectly constitutional measures or what looked to the outside like perfectly constitutional measures. They were able to little by little lockdown dictatorship in their hands.

But if all you looked at was the form, you wouldn’t have realized that dictatorship was coming back in those places. You would’ve thought, “Ah, a new leader is elected and he’s changing the law in order to fulfill his mandate.” So it took quite a lot for other people to catch on. But that leading edge, shall we say, of dictatorship by law turned out to be a transportable model.

You could pack it up and you could move it to other countries. And now that’s the dominant way that democracies fail is through aspirational autocrats being elected in free and fair elections, coming to power, changing the law, usually in illegal fashion, until the next thing you know, there’s no way to get rid of the autocrat.

Katy Glenn Bass:

I know you’ve written about this a lot, this concept of autocratic legalism. And is it fair to say that when you first started describing that or sort of following that story, it was something that was characteristic of new democracies or countries transitioning out of authoritarianism?

Kim Lane Scheppele:

Yeah, those were the ones I focused on first, but then it turned out that it spreads.

Katy Glenn Bass:

Right.

Kim Lane Scheppele:

And in fact, at the beginning, I think a lot of analysts didn’t take this movement all that seriously precisely because it was easy for them to say, “These are new democracies.” They weren’t really established. Easy to knock over. And so therefore, what happens in new democracy stays in new democracies, and longstanding democracies are relatively immune from all that, but now we know that that’s just not true.

Katy Glenn Bass:

So this concept of autocratic legalism and this belief that it was something confined to new democracies, how do you see this now that we’re in 2026? Is this what the United States and other established democracies are beginning to grapple with, this same sense of autocrats manipulating the law?

Kim Lane Scheppele:

Yeah. So I think we’re certainly seeing that in the United States. What we see is that President Trump elected in a free and fair election comes to power and he issued I believe it was something like 325 executive orders in his first year. The only president who had issued more than that was Franklin Roosevelt who was also involved in a constitutional revolution. These same methods can be used whether you like the revolution or not.

And now what we see is those executive orders being litigated up to a friendly Supreme Court that Trump had an opportunity to, shall we say, change the composition of in his first term. And if all of that succeeds, if the Supreme Court goes along with the executive orders over existing statutes and over prevailing interpretations of the Constitution, we’re going to find ourselves with a new Constitution.

Now, let me just say this isn’t something Trump dreamed up on his own. First of all, I think we attribute way too much to him when he’s out playing golf most of the time. This is something that really was a kind of blueprint that came from The Heritage Foundation Project 2025. You can see him just marching along through all of the things Project 2025 identified. But it turns out that Project 2025 was not just an American project.

As it turns out, Viktor Orbán, Prime Minister of Hungary, and his big think tank, the Mathias Corvinus Collegium is the big mothership in Hungary, and the similarities are so strong between what Orbán did to take over his government in 2010 and what Trump is doing in his first year. That when I read Project 2025, the first thing I said was, “Wow, somebody’s really studied Hungary closely.” And sure enough, what we’ve seen is the very same script being applied here.

To fill that in, let me tell you just two things that were crucial in Orbán’s first year. The first thing he did was to come into power. And while he’d been out of power, he analyzed the national budget to figure out what are all the ways this budget supports people who are likely to oppose me. Comes into power and immediately slashes the money for everything. Puts the opposition into disarray. He defunds them.

They’re running around trying to get their money back and not paying attention to what comes next, which is this shoveling through tons and tons of laws. So what happens again, the other thing that Orbán did was to come in, suspend the civil service law, mass fire public employees so that he could then reinstate the civil service law, hire his own cronies into those positions, and have an administrative state that was not going to stand up to him.

So just those two things alone account for a big chunk of the disorientation when Trump came in those first few months. So really the same kind of playbook happening again right here in the US.

Madhav Khosla:

So Kim, one of the things that I think somewhat follows from your important insight is that we are seeing changes through law, but we are also therefore in some ways seeing changes through lawyers. And lawyers in a way are central to the changes that are happening.

And even before these developments became front and center in our lives, you’ve actually spent a lot of time thinking about how lawyers may not actually be the natural allies of liberal democracy. And some of your work has challenged the conventional understanding of what academics refer to as the legal complex and things like that. And could you walk us just a little bit through this?

Kim Lane Scheppele:

Yeah. So first of all, Orbán and Putin themselves were lawyers. A lot of these autocrats that come to power either are lawyers or they have lawyers that they’re working with in close association. And part of that is because, as I put it, I have a chapter title in this book that I’m almost done with, the book, by the way, is called Destroying Democracy By Law, right on point to your podcast.

One of the chapter titles is called Law is the way the State talks to itself. Now, the reason why I use that formulation is that it’s one thing as king to command somebody to do something in particular, but the only way you can command large numbers of people effectively is to give them general orders. And those general orders are what law actually is, particularly administrative law, particularly orders to the administrative state itself.

But also by entrenching ideas in law, it means that you get followers of two kinds. One of the kinds that were already on board with your political program, they do anything you told them to anyway. But the second thing is you get on board the rule followers, the people who say, “Because something is law, therefore we have to obey it.” And any successful democracy should have a large number of people in that second category.

That’s how this whole thing works. And so if you change the law, people may say, “Well, I disagree with it, but it was made by an elected official in a proper process and therefore I am bound by it.” It’s exactly what you want citizens to think. These are good citizens. So when you put into that process laws that actually have the effect of undermining democratic institutions, you get a lot more buy-in than if you simply did it by edict or by force.

And so it legitimates the entire enterprise of undermining democratic institutions. So that’s why law becomes so crucial. Now, who are the lawyers who do this stuff? Some of the lawyers who do this are lawyers who believe in the leader and believe in the program. But oftentimes you get lawyers doing this because law is like a giant math problem if you like math.

It’s just fun to play with the rules and think, well, gee, how would you do this? How would you do that? Is there a way you can make an argument for this, that, and the other thing? And frankly, legal education in many places, and here I include the US, lends itself toward exactly this kind of lawyering, because how often do we take first year law students and say, “Argue the opposite of what you believe? Come up with the best argument for the position that you don’t have."

How often do we say that lawyers in private practice have the highest duty to represent their clients within the boundaries of law, but often making creative legal arguments that would lead to a result that the lawyers themselves would not choose? So yes, we have legal ethics. Yes, lawyers have some boundaries, but think about what legal ethics looks like.

Very few of the rules are about the values that lawyers are supposed to uphold. So yes, they’re supposed to uphold the Constitution, that’s the main values-based rule, but what if the Constitution is changing? What if the new political movement says the Constitution’s something else? You can swear allegiance to a changing constitution and be completely on board with an autocratic project.

So lawyers have learned to be hired guns. They’ve learned to think that clever legal arguments are good lawyering, and they have much less training and much less grounding in the values that underwrite Democratic institutions so that the lawyers themselves may not realize how much they’re undermining Democratic institutions with the very things they’re recommending.

Katy Glenn Bass:

So I think that’s exactly right. And we’ve been thinking a lot about the role of legal education in terms of what young law students are told when they’re in training in terms of what the expectations are and the duty to zealously represent clients. And I’m wondering, are there comparative examples of different ways of educating young lawyers that you know of from other countries or what would you do if you were in charge of reforming legal education in the United States to try to remedy some of those problems, especially in such a polarized time?

Kim Lane Scheppele:

Right. So this is very difficult because in the United States, what we teach is, yes, you are bound to the Constitution. You swear an oath to the Constitution, not to the president, and yet the Constitution itself, how do we know what the Constitution is? And the answer has been, the Supreme Court tells us that. But what if the Supreme Court’s been captured, especially by a faction that isn’t committed to democracy?

The Supreme Court starts issuing cases with which you disagree, but you say, “My oath to the Constitution is an oath to the Supreme Court.” We don’t have a way for lawyers to distinguish between a Democratic constitutional vision and contingent decisions of a court that may have come under political direction. And so this is what you see happening right now in the US.

I have two sets of lawyer friends. One is the set of lawyer friends, and mostly these are people who have international experience because they’ve seen this happen elsewhere. They say, “Wait a second, this is a counter-constitutional project. The point of this is to unsettle and undo the constitutional orders that we have. So therefore, I’m going to go into the resistance and fight to prevent that from happening."

So that’s a small set of lawyers. The other lawyers are saying, “Well, we’re still doing law. We’re still litigating. We’re still pretending like all this is real. Yes, we’re disappointed with Supreme Court decisions, but we’ve learned to live with the ones and then you go back to fight another day.” And everybody is assuming that all the judges up and down the system are still doing law. And that’s the big question.

What happens when you get autocratic capture? One of the first institutions the autocrats go after are the courts precisely to take advantage of the fact that lawyers are trained to follow them to legal education. So then how do we train lawyers elsewise? This is where it seems to me we have to be thinking about courses in democratic and constitutional theory that are not tied to court decisions, that are all about the point of the enterprise.

So let me just tell you one course I used to teach when I was a law professor. I was a law professor at Penn for a decade. We were trying to reform legal education as every law school keeps doing. We decided that students were reading too many cases in their first year, that they needed a break from that. We had an interdisciplinary faculty, so we put in place this requirement that students had to take one elective that would get their nose out of case law.

So I decided to offer a course, which was a role play of Madison’s notes on the Constitutional Convention. And if you read that, it’s a very dense... It’s like reading the phone book, frankly. You have to really push yourself through it because it’s just a lot of keeping track of who said what and exactly what the votes were, and it’s really hard. Turns out if you do it as a role play, take it as a script for a play and you assign students to play all the different parts.

So somebody’s Elbridge Gerry and somebody’s Ben Franklin and somebody’s James Madison, et cetera. And then what happens is the book comes alive. So what I would do with my students, I’d get them to take up these various roles. Then I’d have them go off and read biographies of their characters, read something about their states, how they were elected to serve in the Constitutional Convention.

And then each person had to keep track of what their character said through the debates and we kind of reenacted them in class. And then when we got through the whole thing, these delegates had to go back to their states and work their way through the ratification conventions and report back to the class on what happened in the ratification conventions.

So I thought of it as my anti-originalism course, because once you’ve done that, you realize there were so many different views on offer, so many different debates, that the idea that the framers were one kind of thing just is belied by a simple reading of all of this. But I realized in retrospect that part of what we were doing in that class was really asking the question, what should constitutions do?

And I think it empowered students both to challenge even the Supreme Court’s version of originalism now, but to recognize that in any robust constitutional democracy, you’re going to have a lot of different views on offer and some of them lead down the path to dictatorship and some of them allow you to maintain your democracy. So that was just one example of a course that you could teach that doesn’t put the Supreme Court in the middle of what the Constitution means, gives people resources to fight back.

Madhav Khosla:

One of the other things that I think that is sort of worth perhaps talking about is that the legal profession is self-regulated like the professions are in modern societies. And how effective do you think that self-regulation has actually been in curbing the problem in the countries that you’ve studied?

Because you’re absolutely right that we don’t really have a system where lawyers are taught to distinguish between ordinary democratic moves and a kind of democratic backsliding. I was just wondering if, for instance, in Hungary, has the bar pushed back at all against Orbán or the lawyers who’ve supported his efforts? Or for example, have you seen actually the bar refusing to represent people who might be against the government?

Kim Lane Scheppele:

Well, unfortunately, no, not in Hungary, not in Russia. They all caved. And part of this is because if what you’re doing as a lawyer is winning cases in court and the courts come to be captured, fighting back against the system just guarantees that your clients are going to lose, right? So you need the bar to have a bigger field of vision than just representing their clients.

Occasionally what’ll happen is you’ll get in some places the bar pushing back. So for example, in Poland, the bar pushed back. This was a pro-autocratic government came in in 2015. They immediately started attacking the judiciary. One of the things that they were attacking was, among other things, the role of the bar in picking judges, which they diminished. But it was also the case that they were using unlawful methods to put judges onto the constitutional tribunal.

The bar actually stood up to that, as did a number of judges. And so what you’d see are these lawyers marches where everybody would show up in their legal robes, the kind of things they’d appear in court in, and they would march in the streets with signs saying, “Save our judges.” It was really quite moving. And when the president of the Supreme Court was fired, she insisted that she’d been fired unlawfully.

She decided to go to work as if she had not been fired. Hundreds of lawyers showed up to escort her into the courthouse. So there are moments when you see the bar really rising to the occasion.

Malcolm Brabant:

Free courts now is the Clarion cry. Outside a courthouse in Central Warsaw, demonstrators demand the removal of a judge appointed by the populist conservative government to replace one of a more independent spirit. They accuse the country’s justice minister of being a judicial puppet master.

Michal Wawrykiewicz:

We are still on the battle for the rule of law in Poland. The rule of law is...

Kim Lane Scheppele:

One thing that I think helped that in Poland was the fact that Poland is part of the EU. There are now sort of EU wide bar associations, so that if your country is coming under attack, the next country over may not be, and you get sort of an echo back from people outside your system about what your system looks like to others. And so there were a number of these bar groups that then went to court to challenge the fact that the European Union was not doing enough to preserve judicial independence in Poland.

So again, it was lawyers associations across Europe that did that. Now, it helped that Poland was the second case. Hungary was the first. When the Hungarian judiciary was being compromised, it was so novel people didn’t know what they were seeing. But just to bring this back to the US, one of the interesting things that happened here well before Trump came back the second time, I’ve been going around talking to different bar associations because they all think all these comparative examples are kind of interesting.

And one thing I started doing in some of these bar association talks was to go in and look at what big law did in those countries when the autocratic consolidation started. And the answer was that they turned tail and ran. They often turned over their legal business to local lawyers and they just shut down their offices. So here’s the thing, big law was not brave in other countries, these big international law firms with thousands of lawyers with offices in every major capital.

When the autocratic impulse came over these rulers, the big law firms hung in there for a while, and then they left. What do we expect them to do in the US if that’s what they did abroad? And I’m a little surprised because I would talk to these bar groups and literally name the firms that left Hungary, that left Poland, left Russia, et cetera, and no response. It was so interesting.

So from that, I guess I wasn’t surprised to see what’s happened here, which is yes, Trump singled out law firms with executive orders and some of them caved and some of them fought. That’s usually the story. But what really happened was that all the law firms, including the ones that fought, have really avoided getting into the trenches, fighting against Trump’s executive orders.

They’ll still defend their clients even if their clients are leaned on by Trump, but they won’t put their necks out and actually start to figure out how to use law creatively to push back against the constitutional revolution that we’re seeing. Big law is just all out.

And the bar associations, when they get complaints that, for example, that there are DOJ prosecutors who have violated basic legal ethics, you can actually list the numbers of the rules that they violated, the bar associations just say, “Oh, we’re going to wait for a criminal procedure before we look at what our lawyers have done,” or they start an endless investigation that goes down a rabbit hole and never comes out.

So there’s a couple of bar associations that have disbarred people who have obviously acted in violation of their oaths, but not very many. So the US bar is, shall we say, protecting itself and not protecting the Constitution.

Katy Glenn Bass:

That’s really interesting. I didn’t know that story of the big law firms in the other countries. Can I ask, I mean, how different or not is the situation of US civil society where the bar in the US, lawyers like Knight Institute lawyers where we are bringing cases from a more independent neither government nor a big law stance, is there a difference in terms of the size and the relative wealth of US civil society versus some of these other countries? And do you think that has had any meaningful difference?

Kim Lane Scheppele:

Yeah. So one of the things about established democracies as opposed to new ones is that the established ones tend to have bigger and deeper civil society organizations. And so what you see in the US is Knight Institute, Democracy Defenders, Democracy Forward, Public Citizen, Protect Democracy, ACLU, and Democracy Docket, Marc Elias, from doing all the voting stuff, what you see is there’s a good half dozen to dozen groups like that that are carrying the heavy load of fighting back in court because the lower court judges are still holding the line even if the Supreme Court’s not.

And it’s worth litigating, it’s worth fighting this stuff just to throw sand in the gears, slow things down. We forget that there’s about 150, 160 injunctions that are still holding out there. A recent report showed that of all those big science grants that were cut at the beginning of 2025, by the end of the year, 80% of them have been restored through litigation. So litigation is still an important path.

And it’s these NGO interest group, mid-sized, overworked, stressed out groups of lawyers that are handling the huge fight. I’m struck by the big difference between what’s happening now and what happened after 9/11. So after 9/11, there was a lot of overreaction on the part of the Bush administration. There was some civil society pushback, but then the torture memo came out.

Then it was apparent that the US government was aiding and abetting a program of what they were calling enhanced interrogation. There were renditions and detentions that were being done in violation of basic due process norms of international law. And so as soon as that came out, it turns out that big law was all in. Every single firm, from left to right, devoted pro bono work to torture victims, to legal detentions, to those rendition cases.

I know because I actually was running a LISTSERV for all the lawyers that were involved in all that work to pushing back against the Bush administration and the war on terror. It could not be more different now. Big law has just gone silent. What you see now are partners that are quitting. Some of them are forming their own firms, but they’re small firms. They don’t have the immense deep pockets of big law.

So I feel like from a legal standpoint, we’re fighting the onset of dictatorship in the United States with one hand tied behind our back by lawyers who insist upon thinking that we’re still in some kind of normal politics. And I’m very worried about that. If the bar isn’t there to defend the Constitution, the legal order, basic rights, due process, then what’s the average person to do who doesn’t have the skillset to even know how to name those things?

So yes, I’m a little disappointed in the organized bar. We need to do a whole lot better than that if we’re going to save democracy and lawyers are going to be crucial to that, even if they can’t do it alone.

Madhav Khosla:

Kim, do you think some of that potentially has to do with the fact that a lot of people in big law don’t see themselves as lawyers in the way that you and I conceive of lawyers? Maybe actually like the people who you mentioned in Hungary, maybe they just see themselves as basically figures engaged in transactions rather than your Blackstonian advocate who’s meant to stand up for things.

I mean, maybe it’s not fortuitous that the firms that actually stood up against Trump are those with large litigation practices. And so actually a lot of what you and I think of as big law is basically just bankers in disguise or something like that. I mean, I’m obviously putting the point provocatively, but I was wondering whether there’s something about the changing character of the profession itself that captures some of this dynamic.

Kim Lane Scheppele:

Yeah. What you say is what actually quite a few people have been arguing, that basically lawyers are doing much more than law and they’re doing transactional work. They’re arranging their client’s finances. They’re engaged in preventative action to keep them from ever being sued. They’re organizing their paperwork essentially. So yes, that happens. But here’s what’s really interesting, it’s until gravity disappears, you don’t notice how much you rely on it.

And transactional work relies on there being a fundamentally rule of law based system that underwrites all those transactions. There’s a wonderful book by a legal anthropologist, Annelise Riles, and she studied the Japanese Stock Exchange. And what she studied, and if you think about it, unless you’re in this line of work, you never think about this problem. If you buy and sell a stock, suppose you buy a stock, somebody had to sell it.

But when you buy and sell in an exchange, you have no idea who the seller was when you buy stock. Somewhere underneath all that, there’s a whole bunch of lawyers writing contracts. And Annelise studied the backroom of the Japanese Stock Exchange and all the people, all the lawyers who were matching the buys and the sells to figure out how to do the legal contract that underwrote the exchange.

Now, you don’t see those legal contracts unless there’s some problem, but they have to be there so that there won’t be a problem. And a lot of the legal system operates invisibly like that, where it just is underwriting everything we count on. When I teach sociology of law, I teach what is marriage, what is a passport, what is money, what is Princeton University? They’re all legal forms.

They’re not something that exists apart from the legal guarantees that underwrite them. So you could be a transactional lawyer thinking that all you’re doing is just rearranging the subsidiary over here and moving assets over there and setting up some kind of firewall so that a suit won’t deprive the firm of all of its resources over here. All that stuff. But all of that stuff may not take you into a courtroom, may not take you near the government.

But if you don’t have the infrastructure of courtrooms and governments and basically the guarantee of the rule of law that a democratic government provides, all those transactions are simply paper. And I just wish lawyers realized they rely on think of it as legal gravity that holds everything down and keeps it from floating off into the air. They can’t escape that and they have a responsibility for protecting that.

Katy Glenn Bass:

Do you have any sense that lawyers in the US are becoming more aware of that legal gravity or sort of becoming more aware of that fundamental underpinning of the rule of law that they need? Is there a tipping point that we should expect?

Kim Lane Scheppele:

Well, yeah, I think we’re seeing it a little bit. I mean, so first of all, when Trump first came in, things happened so fast and people couldn’t believe what they were seeing, that they preferred not to see it. And I think now the destruction, the rubble, the Justice Department lies in rubble. Trump’s picture appears on the Justice Department. I mean, there’s so many signs that our legal order is not in good order.

There’s so many signs that we’re getting from the Supreme Court that the Supreme Court doesn’t know what the law was, or rather it doesn’t care what the law was. It will do something else. I think if you’re paying attention, what you realize is we’re very far into autocratic capture of our legal system and we’re very far into the destruction of the very institutions that will allow peaceful transfers of power in the future. Now, if people realize that...

I mean, one of the problems is that by the time you realize it, it may be too late to fix it, right? So you’re then in this situation where all you can do is patch the rubble back together again somehow. So I think we’re already so far along in the destruction of democratic institutions, but we’re now at the point where you have to realize we’re not just going to pick up the rubble of the Justice Department and try to patch it back the way it was.

We’re going to have to reimagine the kind of thing the Justice Department is. We’re not going to just be able to go to the Supreme Court and say, “Cut it out. Follow your own precedent.” We’re going to have to reimagine what a Supreme Court will look like. I mean, this is really major reconstruction, I think. So just for example, what most countries do that have been through dictatorships, one of the first things they do is they separate the Justice Ministry from the prosecution service, and they wall off the prosecution service from political influence.

Wouldn’t that be a good idea for us? We just happen to have this Justice Department because we were the first comer and all those things are packaged into one institution, but most governments separate them. There’s also another thing which is that most governments don’t have judges that stay on the bench for 30 years at their highest court level. So they have term limits.

They have age limits. They have processes that guarantee that the judges who get appointed, especially to the highest court, are named in procedures that require cross-party agreement. You could start imagining importing all these ideas that other countries have come up with. I mean, America hasn’t been great at borrowing back because we think we invented it.

But it’s like we’re driving around in a Model T Ford and everybody else has got these fancy cars that can do things that Model Ts couldn’t do. And we’re saying, “But isn’t the Model T wonderful because it came first?” Constitutional drafting has gone way past where the US is. And we’re now so far into this system of autocratic capture that one of the things I do is I teach about all these other systems.

And I left law school teaching in order to teach the undergraduates before they get to US law school. So that when they get to US law school, they’re going to say, “Really? That’s how we do things? There are all these other ways you could do things. Why do it like that?” So I’m being a subversive by getting to the undergraduates first, but I think we need to look around a little bit.

Katy Glenn Bass:

Kim, is there anything else that’s on your mind on these topics that you know so well that we haven’t touched on yet?

Kim Lane Scheppele:

Yeah, so I actually didn’t lean very much into the self-regulation question. It’s the same issue that’s now coming up with regard to academic freedom. When you have a self-regulating profession, it actually has to take self-regulation seriously. And bar associations have been sort of polite clubs. They haven’t wanted to set and enforce the rules very seriously. Yes, there are disbarment procedures.

It’s usually in egregious cases by lawyers who aren’t that powerful. You’re not seeing very powerful lawyers doing very big things getting disbarred, right? And we have the same problem with academic freedom that we’ve allowed things to happen in universities that probably shouldn’t have happened if we had been policing standards of not just disciplinary conduct, but of what universities are supposed to be for.

So we have to think about self-regulation as not just are the I’s dotted, T’s crossed in particular behavior of those who are regulated. We have to think bigger and more theoretically about the principles that we’re supposed to be defending. And there’s a reason why the bar has always been thought of as separate from ordinary jobs. It’s because it’s a normative enterprise as well as a practical one.

Americans used to think, if we swear an oath to the Constitution, that takes care of itself. What if the Constitution breaks? Then what? And we’re sort of in that moment when just following decisions of the Supreme Court as an oath of office, it may come... I should say this, this is my revolutionary statement, there may come a time when the patriotic and constitutional thing to do is to disobey a decision of the Supreme Court.

I hate to say that. Other countries have gotten to that point. You can’t just do it as an act of force and violence. You have to do it because you’re invoking constitutional principles that are more important and more enduring. And if we have no idea what those are, we don’t know what the point is at which we have to say, “Just having the Supreme Court tell us what the Constitution means is not enough."

And that’s what I wish the bar would think about. It’s a moment when lawyers should be among the bravest because the tactics that are being used to dismantle democratic institutions are legal ones. The lawyers will see what’s happening before everybody else.

We need to be the early warning signal to everybody else that this is what’s going on, and we need to play a key role in building these things back, because ultimately a government of laws is a government of laws and lawyers do have some privileged position in figuring out, designing and enacting those laws. And so where’s the bar? We need the bar.

Katy Glenn Bass:

This has been such a wonderful conversation, Kim. Thank you so much for making time to talk to us today and for sharing all of your insights from your years of scholarship and advocacy.

Kim Lane Scheppele:

Well, thank you so much for doing the whole podcast. And I really hope you get a lot of listeners.

Katy Glenn Bass:

Us too.

Madhav Khosla:

Thanks so much, Kim.

Katy Glenn Bass:

Lawyering Without Law is a production of the Knight First Amendment Institute at Columbia University and is hosted by Madhav Khosla and me, Katy Glenn Bass. This episode was produced and engineered by Dustin Foote. Fact checking by Connor Menzies and Sofia Rojas. Candace White is our executive producer. Our music comes from Envato Elements. The art for our show was designed by Jay Vollmar.

The archival clip you heard early on in the show was from PBS. Thanks to Kim Lane Scheppele who had joined us for this episode. Lawyering Without Law is available on Apple, Spotify, and wherever you listen to podcasts. Please subscribe, share, and leave a review. We’d love to know what you think. To learn more about the Knight Institute, visit our website, knightcolumbia.org.

That’s Knight with a K. And follow us on social media. We’ll see you next time for a conversation with Harvard Law Professor Lawrence Lessig. Bye for now.

I. Introduction

Transformative ideas and transformative communicative practices from around the globe have enriched American public discourse and updated how Americans interact with one another. Social media platforms are only the latest means of communicating and engaging. However, foreign-owned social media platforms that offer novel, participative ways for people to engage, communicate, and associate are a form of communications infrastructure that is not adequately protected under US law. Although there is a focus on platforms’ communicative features, there is very little discussion of their associational features.

The Supreme Court has recognized Americans’ right to access foreign ideas.1 1. Lamont v. Postmaster General, 381 U.S. 301 (1965); See also Xiangnong (George) Wang, Listeners’ Rights in the Time of Propaganda: The Story of Lamont v. Postmaster General, Knight First Amendment Institute, http://knightcolumbia.org/content/listeners-rights-in-the-time-of-propaganda-the-story-of-lamont-v-postmaster-general (last visited May 6, 2025). This right is sufficient to ensure access to mass-media forms such as print and broadcasting that depend largely on one-way transmission to audiences.2 2. Although the mass-media also encouraged engagement in the form of letters to the editor and telephoning to engage with radio and television shows, this formed a very low percentage of content consumed by the audience in comparison with social media. But it is insufficient to regulate participatory social media platforms. Together with mass media, but increasingly dominating it, social media forms the underlying infrastructure of communication in American society.3 3. See Robert C. Post, Data Privacy and Dignitary Privacy: Google Spain, the Right to Be Forgotten, and the Construction of the Public Sphere, 67 Duke L.J. 981, 1070–1 (2018) (describing Google as "communicative infrastructure"); Brett M. Frischmann, Infrastructure: The Social Value of Shared Resources 4 (describing infrastructure as "shared means to many ends") (2012). Accessing this infrastructure is distinct from, although connected to, the right to receive ideas. The right to communicative apps as infrastructure is the right to access new ways of communicating and associating. Legal discourse needs to build on the right to access foreign ideas, extending it to the right to access foreign communicative infrastructure.4 4. See Part 6 below; see also Lamont v. Postmaster General, 381 U.S. 301.

My goal in this essay is to reframe the value of foreign social media platforms (and potential future AI platforms) in the context of TikTok Inc. v. Garland as unique and creative associational and speech infrastructures and show why associational and speech infrastructures should be treated differently from other foreign infrastructure.5 5. I am responding in part to scholars who argue in favor of treating speech infrastructure like banking or energy infrastructure. See e.g., Ganesh Sitaraman, The Regulation of Foreign Platforms, 72 Stan. L. Rev. 1073 (2022). The justices adopted an outdated view of the significance of foreign social media—suggesting that a platform’s users are the only people affected by the platform—which led them to underestimate the burden on speech from restricting the app or compelling a change in its ownership. Since lawmakers and experts, including TikTok’s advocates, share this outdated view, the TikTok case is an excellent opportunity to discuss foreign platforms’ role in the American public sphere.

My argument for access to foreign communicative infrastructure extends to all significant foreign communicative apps, including chatbot-centric apps. Political, economic, and cultural influences shape communicative apps, and each app amplifies different kinds of content and offers different ways to engage. Changing foreign apps’ ownership removes these influences and burdens speech in ways that judges should take into account. Some of these questions were discussed in the context of the ‘TikTok ban’ in 2024 and 2025. Although the TikTok question has been settled, TikTok Inc. v. Garland is worth discussing because it presents the Supreme Court’s view of the role of foreign communicative apps in the American public sphere.6 6. TikTok Inc. v. Garland, 604 U. S. 56 (2025).; The Supreme Court evaluated the Protecting Americans from Foreign Adversary Controlled Applications Act in the context of an as-applied challenge brought by TikTok, and the justices clarified in their per curiam opinion that their “analysis must be understood to be narrowly focused in light of these circumstances.” Nevertheless, the case is indicative of how the Supreme Court applies the First Amendment to laws regulating foreign social media platforms. It appears that the Supreme Court will permit Congress and potentially the president to restrict Americans’ access to Chinese-made apps. Unfortunately, the judgment's implications for social media regulation have been eclipsed by the events that followed it. The justices’ application of the intermediate scrutiny test was based on a dangerously flawed understanding of social media.7 7. I am not discussing the question of whether the strict scrutiny test should have been applied since others have already done so. See e.g., Anupam Chander, Gautam Hans & Edward Lee, TikTok v. Garland Opens the Door to Global Censorship, Lawfare (2025), https://www.lawfaremedia.org/article/tiktok-v.-garland-opens-the-door-to-global-censorship. I focus on the meaning and consequences of the TikTok ruling for the American public sphere, not the actual TikTok divestiture.8 8. Executive Orders, Saving TikTok While Protecting National Security, The White House (Sept. 25, 2025), https://www.whitehouse.gov/presidential-actions/2025/09/saving-tiktok-while-protecting-national-security/.

II. Communicative Apps as Infrastructure

Infrastructure such as language, script, the press, and broadcasting make human communicative activity possible. Such infrastructure increases shared knowledge and shapes public opinion.9 9. Post, supra note 3 at 1020-1023. Online communicative platforms extend this ability by enabling Americans to share, modify, and consume content in conversation with each other and people around the world. In so doing, the platforms facilitate public discourse, including discourse on other platforms and in the media.

As “shared means to many ends,” most communicative apps are infrastructure.10 10. See Frischmann, supra note 3 at 4. Although other legal scholars have made this argument,11 11. K. Sabeel Rahman, The New Utilities: Private Power, Social Infrastructure, and the Revival of the Public Utility Concept, 39 Cardozo L. Rev. 1621 (2017); Julie E. Cohen, Between Truth and Power (2019); Julie E. Cohen, Infrastructuring the Digital Public Sphere, 25 Yale J.L. & Tech 1 (2023); Evelyn Douek & Genevieve Lakier, Lochner.Com, 138 Harv. L. Rev. 100 (2024). more work is necessary on what this means for the sectors these apps serve as infrastructure. In this essay, I discuss the implications for speech of treating significant social media platforms as communicative infrastructure. I use the term “communicative infrastructure”12 12. See Gabriel Tarde, The Public and the Crowd (1901), in Gabriel Tarde on Communication and Social Influence , 52–55 (1969); Post, supra note 3 at 1016, 1046. to refer to both the media ecosystem as a whole and its individual parts (e.g., The New York Times, the BBC, TikTok, Google Search, and Google’s AI Overview). However, my focus is on the digital public sphere, which consists of layered infrastructures that are regulated differently, including telecommunications and the open internet.

Legal scholarship on social media’s infrastructural features tends to focus on how platforms are shaped by the market, political institutions, legal frameworks, and public policy, and how they use the data they collect.13 13. See e.g., Sitaraman, supra note 5; Zephyr Teachout & Lina Khan, Market Structure and Political Law: A Taxonomy of Power, SSRN Journal (2014), http://www.ssrn.com/abstract=2490525. It facilitates a granular discussion of platforms as infrastructure, and specifically of how platforms’ infrastructural role enables public discourse by affecting which people communicate with each other and how they communicate.14 14. Post, supra note 12 at 1016, 1046 (arguing that platforms are communicative infrastructure). I aim to develop the speech-enabling aspect of online communicative platforms’ infrastructural role. Treating communicative apps as infrastructure implies recognizing that these apps have downstream effects beyond the platform.15 15. Frischmann, supra note 3 at 63–5 (discussing how the social benefit of infrastructure derives from its downstream uses). My goal is to contribute to First Amendment scholarship while integrating the valuable insights that scholars of law and political economy offer.

Communicative apps are “structured arrangement[s] that facilitate human [communicative] activity across space and time.”16 16. I have adapted Professor Julie E. Cohen’s definition of infrastructure for communicative infrastructure. See Cohen, Infrastructuring the Digital Public Sphere (2023), supra note 11 at 4. For a discussion of platforms and infrastructure, see Cohen (2019) 40-41. In addition to focusing on the influences on and design of these arrangements, and their harmful downstream effects, it is worth focusing on the extent to which they enable democratic goals.17 17. See Cohen (2023), supra note 16 at 4 (Cohen points out that data collection and targeted advertisement is among the key goals of these apps, which I am in agreement with. My modified definition is meant to highlight the downstream activity that should be significant for First Amendment scholars and lawyers).Communicative infrastructure, as well as the absence of state censorship, is necessary for free speech, including public opinion formation and debate.18 18. Jack M. Balkin, Digital Speech and Democratic Culture: A Theory of Freedom of Expression for the Information Society, 79 N.Y.U. L. Rev. 1, 52–55 (2004). Many democratic constitutions restrict the regulation of the press because the press is a type of communications infrastructure, and such infrastructure supports public discourse.19 19. Post, supra note 3.

Widely used communicative products, such as Facebook and TikTok, have additional infrastructural characteristics. They are embedded in other social arrangements and technologies, they invisibly support tasks, they reach beyond single events and practices, new participants acquire familiarity with them by using them, they have a standardizing influence on other infrastructures and tools, and they become visible upon breakdown.20 20. Susan Leigh Star & Karen Ruhleder, Steps toward an Ecology of Infrastructure: Design and Access for Large Information Spaces, 7 Information Systems Research 111 (1996) republished in Bowker et. al. (ed.) Boundary Objects and Beyond, 380-381(2015); Cohen (2019), supra note 11, at 235. These elements are also increasingly found in chatbots that rely on large language models and AI-based creation tools embedded in social media apps.

III. Infrastructure is Different from Ideas

As Jack Balkin has pointed out, the First Amendment prevents state interference with speech, but infrastructure is necessary to ensure that individuals have the freedom and ability to participate in public discourse.21 21. Jack Balkin, Cultural Democracy and the First Amendment, 10 Northwestern Law Review 1053, at 1060, 1078 (2016). Consider the infrastructure required for public discourse: to access materials and inspiration, to consult others, and to circulate ideas and conclusions. Communicative infrastructure—extending from the press to generative AI—is critical to maintaining public discourse through its structural role in disseminating information.22 22. See Rahman, supra note 11 at 1670–1. This is different from and at least as important as the information itself. Along with the right to access ideas, one needs the means to access ideas.23 23. See generally James Grimmelmann, Listeners’ Choices, 90 U. Colo. L. Rev. 365 (2019).

Furthermore, communicative infrastructure exerts an influence beyond its own systems.24 24. Frischmann, supra note 3 at 337-9. When the state restricts the speech of, say, a significant university or the operations of a major social media platform, more people are affected than those directly involved with the university or platform. Universities and platforms, in different ways and for different reasons, make certain associations and speech salient. Major social media platforms influence what stories the traditional media carry, how politicians campaign, and even how democratic institutions share information with the public.25 25. Joan Donovan & danah boyd, Stop the Presses? Moving From Strategic Silence to Strategic Amplification in a Networked Media Ecosystem, 65 American Behavioral Scientist 333 (2021); Robyn Caplan & danah boyd, Isomorphism through Algorithms: Institutional Dependencies in the Case of Facebook, 5 Big Data & Society 1 (2018); Madison Malone Kircher, How Harris, Walz and Trump Are Finding Their Way to Your TikTok Feed, The New York Times, Aug. 30, 2024, https://www.nytimes.com/2024/08/30/style/tiktok-creators-harris-walz-trump.html; Liam Scott, White House to Open Media Access to Podcasters, Influencers, Voice of America (Jan. 28, 2025), https://www.voanews.com/a/white-house-to-open-media-access-to-podcasters-influencers/7953761.html. Other platforms change their practices to compete.26 26. Sarah Jackson, Mark Zuckerberg Told Meta Employees to Zero in on Video Because They’re up against an “unprecedented Level of Competition” from TikTok, Report Says, Business Insider, https://www.businessinsider.com/mark-zuckerberg-meta-faces-unprecedented-competition-from-tiktok-focus-reels-2022-2 (last visited Apr. 7, 2025).

The infrastructural lens illuminates important elements of the relationship between speech, association, and communicative online apps, including social media platforms and AI-powered chatbots. Communicative online apps play a dual role, enabling communicative relationships at an unprecedented scale while also influencing and shaping them in ways that may impede democracy. For example, social media platforms can change what content is available, ranging from single items of content to individual accounts and even whole feeds, and can use removal, amplification, downranking, and shadow banning to shape the kinds of content that are available.27 27. Cohen, supra note 11 at 40 (Collective or[external] harms that inhere in patterns of activity manifesting at scale can be particularly difficult to name, understand, and counteract, and this is doubly true for collective harms entrenched infrastructurally); See also Rahman, supra note 11 at 1672. Through repeated nudges, platforms also influence who is put into conversation with whom.

A single entity can play a greater infrastructural role if it dominates the market, as Google does. When Google’s products are unavailable, the public notices.28 28. Stacy Liberatore, Google Hit with Outage That Plagued Thousands of Users, Mail Online (2024), https://www.dailymail.co.uk/sciencetech/article-13735777/Google-Search-engine-YouTube-outage.html. That is a feature of infrastructure—it is seamless and invisible when it works and visible when it stops working. Many people rely on Google. Courts are increasingly finding that this reliance gives Google too much infrastructural power.29 29. See, e.g., United States v. Google LLC 778 F.Supp.3d 797 . See also Rahman, supra note 11; Lina Khan, The Separation of Platforms and Commerce, 119 Colum. L. Rev. 973 (2019); Amy L. Stein, Rejecting Public Utility Data Monopolies, 112 Cal. L. Rev. 1321 (2024). This is important because not all forms of communicative infrastructure may serve democratic values. Non-state actors can also erode infrastructure, which is why structural regulation, such as anti-trust laws, is necessary to prevent private capture of communicative infrastructure that would stunt the diversity of opinion and group formation.

Social media has a powerful effect on public discourse because it steers communicative relationships. Recognizing this associational power and interpreting the First Amendment to account for it is critical for a healthy public sphere. Up to and including the era of broadcast media, regulating media to maximize its benefits to democracy was seen as consistent with First Amendment values, but the standards changed before the social media cases reached the Supreme Court.30 30. See Twitter Inc. v. Taamneh, 598 U.S. 471 (2023); Murthy v. Missouri, 603 U.S. 43; Moody v. NetChoice, LLC, 144 S.Ct. 2383 (2024); TikTok Inc. v. Garland, 604 U. S.; Jack M. Balkin, Moody v. NetChoice: The Supreme Court Meets the Free Speech Triangle, Sup. Ct. Rev. 1 (2024); Douek and Lakier, supra note 11; Genevieve Lakier, The TikTok Ban and the Limits of the First Amendment, LPE Project (Jan. 22, 2025), https://lpeproject.org/blog/the-tiktok-ban-and-the-limits-of-the-first-amendment/. Veering away from the idea that communicative infrastructure can be regulated to enable and protect First Amendment interests has left policymakers and judges without theories of how to treat foreign communicative infrastructure. Specifically, treating ideas as medium-agnostic and treating media as ownership-agnostic ignores how far ownership affects the manner in which infrastructures shape speech and association. If foreign ideas are valuable, so are foreign ways of relating and communicating.

IV. The Special Role of Foreign Communicative Infrastructure

Apps shape and are shaped by the communities they engage with, the markets they compete in, the organizations that build them, their access to resources, and local laws, among other things.31 31. Id.; Zeyi Yang, How Chinese AI Startup DeepSeek Made a Model That Rivals OpenAI, Wired, https://www.wired.com/story/deepseek-china-model-ai/ (last visited Feb. 1, 2025); Susan Leigh Star & Karen Ruhleder, Steps toward an Ecology of Infrastructure: Design and Access for Large Information Spaces, 7 Information Systems Research 111, 380-381 (1996). Tellingly, WhatsApp was co-founded by a man whose Soviet origins taught him the importance of encrypted communication.32 32. David Rowan, The inside Story of Jan Koum and How Facebook Bought WhatsApp, Wired (May 1, 2018), https://www.wired.com/story/whats-app-owner-founder-jan-koum-facebook/. The Chinese company DeepSeek’s large language models were shaped by resource constraints; the company’s engineers innovated in response to limited access to the powerful new generation of Nvidia chips that power generative AI.33 33. Yang, supra note 31. Local dynamics in China trigger innovation in different directions. This produced ByteDance’s platform Douyin’s evolving business practices and addictive algorithm for sharing and watching short videos. China has a highly competitive short-video market. ByteDance’s platform for international users, TikTok, has successfully tailored learnings from Douyin for international markets.34 34. D. Bondy Valdovinos Kaye, Xu Chen & Jing Zeng, The Co-Evolution of Two Chinese Mobile Short Video Apps: Parallel Platformization of Douyin and TikTok, 9 Mobile Media & Communication 229 (2021) 230-1.

As the spread of the short-video format and TikTok’s practices to engage users to other popular communicative platforms shows, foreign innovation can revitalize the infrastructural practices of American communicative apps.35 35. Mark Zuckerberg Is Blowing Up Instagram to Try and Catch TikTok, Bloomberg.com, May 25, 2022, https://www.bloomberg.com/news/features/2022-05-25/facebook-copies-tiktok-app-to-make-instagram-cool-to-teens; Amani Bayo, How Social Media Influencers Make Money on TikTok, Instagram and YouTube, The Columbus Dispatch, https://www.dispatch.com/story/entertainment/local-celebrity/2025/06/22/how-to-make-money-on-social-media-tiktok-instagram-youtube/82641137007/ (last visited July 24, 2025). Even American social media users who have never engaged with any TikTok videos are affected by changes that American platforms make as they innovate to keep up with ByteDance.36 36. Liv McMahon, Instagram Revamps Algorithm in Battle for TikTok Creators, BBC, May 3, 2024, https://www.bbc.com/news/articles/cmm3yn4pr17o; Emma Job, Instagram Is Considering Creating a Reels App to Compete with TikTok, Tech Edition (Feb. 27, 2025), https://www.techedt.com/instagram-is-considering-creating-a-reels-app-to-compete-with-tiktok. The Chinese company is a major competitive presence in a highly concentrated market and triggers changes in the social media ecosystem that can benefit public discourse as well as individuals.37 37. Social Media Fact Sheet, Pew Research Center (Nov. 13, 2024), https://www.pewresearch.org/internet/fact-sheet/social-media/ (showing a sharp rise in the percentage of adults who use TikTok). Admittedly, these changes such as increased addictiveness can also be harmful, but it is difficult to separate out the harm from the benefits just as it is difficult to separate bad ideas from good ones.

Forcing changes in ownership and making foreign social media platforms American-owned changes the nature of communicative apps in ways that affect speech.38 38. It is too early to be certain but the TikTok deal may have avoided this outcome by licensing the Chinese algorithm which may offer it access to Chinese updates but the licensing has been criticized by those who are opposed to Chinese ownership of social media platforms widely used in the United States. See Associated Press, Larry Ellison’s Oracle Set to Spearhead U.S. Oversight of TikTok Algorithm, Fortune, https://fortune.com/2025/09/23/oracle-tiktok-deal-us-security-algorithm-silverlake-murdoch-dell-trump-2025/ (last visited Jan. 1, 2026); Bobby Allyn, TikTok Signs Deal to Give U.S. Operations to Oracle-Led Investor Group, NPR, Dec. 18, 2025, https://www.npr.org/2025/12/18/nx-s1-5648844/tiktok-deal-oracle-trump. Social media platforms’ design and content choices have an enormous impact on communication. For example, platforms enable and classify participation differently. TikTok’s algorithm recommends third-party content based on how long users engage with videos and react to them.39 39. Yulun Ma & Yue Hu, Business Model Innovation and Experimentation in Transforming Economies: ByteDance and TikTok, 17(2) Management and Organization Review, 382-388 (2021) at 385 (“TikTok touted the algorithm as computer vision to extract and categorize visual information that relies on users’ watching history and engagement patterns to serve users’ interests. In short, TikTok controls your menu of entertainment by observing your reactions to each past video. Therefore, TikTok users don’t need to think and search for the videos but are fed personal preference-based videos, which is a crucial part of TikTok”); Compare with Angela Yang, Instagram and Twitch roll out new TikTok-like short-form video discovery features, NBC News, May 1, 2024 instagram-tiktok-roll-new-tiktok-short-form-video-discovery-features-rcna150231.html (“Instagram announced Tuesday a tweak in its discovery algorithm that would amplify smaller creators through recommendations, which show users posts and reels from accounts they do not already follow through an algorithm that tracks and predicts their interests.”). Those who access this infrastructure have a different communicative experience from those who use other platforms.

A new owner of a communicative app can order communication and association in new ways. The most widely used social media and, increasingly, AI platforms follow an infrastructural model that Julie Cohen describes as “platformized” infrastructure.40 40. See Cohen (2023), supra note 11 at 4, 26; Cohen (2019), supra note 11 at 41. This model enables greater centralized control than is typical for public infrastructure such as a public library system.41 41. Cohen (2019), supra note 11 at 41–2; See also Laura J. Neumann & Susan Leigh Star, Making Infrastructure: The Dream of a Common Language, Proceedings of the 4th Biennial Participatory Design Conference (1996), https://ojs.ruc.dk/index.php/pdc/article/view/153. Platformization places greater control in platform owners’ hands, permitting their ideological orientations and their susceptibility to markets, law, and political power to shape platforms’ decisions about what content and communication they permit and privilege.

Owners’ influence can be relatively subtle on social media platforms and AI-driven communicative platforms because they engage in what I am going to call ‘algorithmic content mediation,’ to distinguish it from editing and curation, both of which imply more precision of control over the output. Newspapers know exactly what they will publish, and their owners have to tell (human) editors what they cannot publish.42 42. Richard Luscombe, Washington Post Opinion Editor Departs as Bezos Pushes to Promote ‘Personal Liberties and Free Markets,’ The Guardian, Feb. 26, 2025, https://www.theguardian.com/media/2025/feb/26/jeff-bezos-washington-post-opinion. In contrast, online platforms can use filters and prioritize some kinds of content over others, but they rarely know exactly what will emerge given the speed and scale of speech online.43 43. Jack M. Balkin, Free Speech versus the First Amendment, 70 UCLA L. Rev. 1206 (2023); Paul Gowder, The Networked Leviathan 46–48, 126–137 (2023).  For example, when Elon Musk started changing Twitter’s (now X’s) algorithms, a menswear writer called Derek Guy ended up being promoted all over the platform for reasons that remain unclear. This is why it is helpful to think of algorithmic content mediation as “deal[ing] in probabilities, error rates, costs, and benefits.”44 44. Balkin, Free Speech versus the First Amendment,70 UCLA Law Review 1206 at 1249.

V. Foreign Ownership and Communicative Infrastructure

By stressing the value of foreign communicative platforms, I am not implying that they should not be regulated.45 45. See Sitaraman, supra note 5 at 1077–8. This essay confines itself to demonstrating that foreign apps contribute more value to the American public sphere than regulatory conversations acknowledge. Regulating foreign ownership of communicative infrastructure should involve assessing how far foreign political and cultural influences enrich American public discourse. Foreign apps, with their distinct influences, offer new ways for people to engage with each other. Infrastructural apps like major social media platforms also evolve continuously. This ought to be taken into account by the Supreme Court when it assesses how far regulating foreign apps burdens speech.

TikTok Inc. v. Garland offers an illustration of what is lost when judges do not account fully for the burden on speech.46 46. TikTok Inc. v. Garland, 604 U. S. __. The Supreme Court acknowledged that the First Amendment rights of American users were implicated in the use of TikTok. However, it underestimated the extent and nature of the relationship between these rights and major foreign platforms like TikTok by failing to account for TikTok’s infrastructural role, and how this role would be affected by transferring TikTok to American owners. In the Supreme Court’s defense, TikTok’s arguments in the case did not highlight its infrastructural role or influence on public discourse and confined itself to discussing how its users were affected.

TikTok enables Americans’ First Amendment rights in three significant, under-considered ways: (1) content from TikTok circulates to other platforms; (2) TikTok, as a foreign platform, offers Americans unique ways to communicate and engage; and (3) TikTok’s practices trigger changes in competing platforms.47 47. See for example, McMahon, supra note 36; Job, supra note 36. Thus, access to Chinese (and other foreign) apps supports Americans’ First Amendment rights.

Lawmakers, judges, lawyers, and others described TikTok as affecting only the rights of its 170 million registered American users rather than accounting for the entire burden on speech.48 48. See TikTok Inc. v. Garland, 604 U.S. __. The enormity of this burden becomes clear when one considers that TikTok affects the speech of more Americans than just its users, who make up a third of all American adults and over half of Americans under 30.49 49. Kirsten Eddy, 8 Facts about Americans and TikTok, Pew Research Center (Dec. 20, 2024), https://www.pewresearch.org/short-reads/2024/12/20/8-facts-about-americans-and-tiktok/. TikTok permits these people to engage with more than a billion people across the world.50 50. Andreas Schellewald, Understanding the Popularity and Affordances of TikTok through User Experiences, 45 , Media, Culture and Society 1568 (2023). It triggers changes in the cultural industries and the social media sector, as other companies, media, and content creators react to TikTok’s innovations.51 51. See Jackson, supra note 29.

As I was writing this essay, President Trump approved a TikTok transfer of ownership agreement.52 52. David Shepardson, US Treasury’s Bessent Says China Has Approved TikTok Transfer Deal, Reuters, Oct. 30, 2025, https://www.reuters.com/world/china/china-us-trade-deal-could-be-signed-next-week-us-treasurys-bessent-says-2025-10-30/. The arrangement did not take the form that most commentators and lawyers—and arguably the Supreme Court—had envisioned it might: namely, that the algorithm would be transferred when the ownership changed.53 53. See Alan Rozenshtein, Lessons Learned from the TikTok Saga, Lawfare (2025), https://www.lawfaremedia.org/article/lessons-learned-from-the-tiktok-saga ("As someone who analyzed and defended the law on both legal and policy grounds from the very beginning, I should be happy with this outcome. And yet I am anything but. I have serious concerns about whether this deal genuinely addresses the national security risks—the opacity of any algorithmic "licensing" arrangement makes verification impossible"). Instead, the TikTok agreement arranged for ByteDance to license the content recommendation algorithm to TikTok’s new owners.54 54. See TikTok’s Algorithm to Be Licensed to US Joint Venture Led by Oracle and Silver Lake, AP News (Sept. 22, 2025), https://apnews.com/article/trump-tiktok-china-d5d8a1d56b5185778536874d7fc1ee62; Rozenshtein, supra note 53. However, I am going to discuss the Supreme Court’s decision in line with the initial expectations that ownership of the algorithm would change, as that allows an examination of the broader question of how ownership affects digital communicative infrastructure.

New owners have a visible impact on the development of platformized apps for the reasons I highlighted in the previous section. This is evident from Meta’s goals for WhatsApp.55 55. Parmy Olson, Exclusive: WhatsApp Cofounder Brian Acton Gives The Inside Story On #DeleteFacebook And Why He Left $850 Million Behind, Forbes.com 4 (2018); Alex Hern & Alex Hern UK technology editor, Twitter Usage in US ‘Fallen by a Fifth’ since Elon Musk’s Takeover, The Guardian, Mar. 26, 2024, https://www.theguardian.com/technology/2024/mar/26/twitter-usage-in-us-fallen-by-a-fifth-since-elon-musks-takeover.Although WhatsApp’s co-founders chose to sell the app to Meta (then Facebook), news reports suggest that they were unhappy with the new owners’ design choices.56 56. Olivia Solon, WhatsApp CEO Jan Koum Quits over Privacy Disagreements with Facebook, The Guardian, Apr. 30, 2018, https://www.theguardian.com/technology/2018/apr/30/jan-koum-whatsapp-co-founder-quits-facebook; Olson, supra note 55. Similarly, Elon Musk’s influence on X, previously known as Twitter, offers an extremely visible example of the impact of ownership changes on communicative platforms, including social media.57 57. See Kate Conger & Ryan Mac, Character Limit (2024), https://www.penguinrandomhouse.com/books/737290/character-limit-by-kate-conger-and-ryan-mac/; Antonio Pequeño IV, Meta’s Threads, Bluesky Gain Users As Musk’s X Dips, Forbes, https://www.forbes.com/sites/antoniopequenoiv/2025/03/25/elon-musks-x-sees-popularity-slide-after-election---while-threads-bluesky-makes-gains/ (last visited June 24, 2025); Dan Milmo & Dan Milmo Global technology editor, ‘Musk Destroyed All That’: Twitter’s Business Is Flailing after a Year of Elon, The Guardian, Oct. 27, 2023, https://www.theguardian.com/technology/2023/oct/27/elon-musk-x-twitter-takeover-revenue-users-advertising; Aisha Counts, Elon Musk Is a ‘Free Speech Absolutist,’ Except at Work, Bloomberg.com, Sept. 14, 2023, https://www.bloomberg.com/news/newsletters/2023-09-14/elon-musk-says-he-s-pro-free-speech-but-fired-twitter-staff-for-comments. A change in ownership means new priorities, values, and goals inform an app’s evolution, which implies that changes in ownership transform communication apps.

Access to a diversity of communicative apps, including foreign apps, is critical to Americans’ right to explore new ways of engaging with each other. The major Chinese platforms are currently the only platforms that offer Americans an alternative to discourse ordered by what is increasingly called the American “tech oligarchy.”58 58. See Julie E. Cohen, Oligarchy, State, and Cryptopia (abstract), dli-cornell-tech, https://www.dli.tech.cornell.edu/seminars/oligarchy%2C-state%2C-and-cryptopia (draft on file); See also Chris Hayes: The Dystopian Vision Trump and His Billionaire Allies Have for the U.S., MSNBC.com (Jan. 8, 2025), https://www.msnbc.com/top-stories/latest/trump-musk-zuckerberg-bezos-oligarchy-rcna186818. Although there are good reasons to regulate foreign apps (and American apps that threaten national security), there is no reason for the regulation to be so expansive that it blocks Americans’ access to them instead of regulating their harmful effects or tendencies narrowly. In this regard, the European Union’s approach to foreign platforms might have been the better approach for American policymakers to adopt. Robert Post has argued that the European Union should treat Google, which is foreign-owned in the EU, as “an essential component of the communicative infrastructure necessary to sustain the public sphere” because of its status as a significant online platform.59 59. Robert C. Post, Data Privacy and Dignitary Privacy: Google Spain, the Right to Be Forgotten, and the Construction of the Public Sphere, 67 Duke L.J. 981, 1016, 1070-71 (2018). This argument could also apply to foreign communicative platforms in the United States: They enable Americans’ First Amendment rights. However, the dominant social media platforms in the United States and most of the English-speaking world have historically been American. This has left American regulatory discourse unprepared for the constitutional questions that foreign platforms raise. Policy aside, the Supreme Court would have been better able to appreciate the scale of speech rights at stake if the problem in TikTok Inc. v. Garland had been presented as a right to access communicative infrastructure. Framed correctly, the TikTok case might have been the Lamont of modern media.

VI. Adapting Lamont for Web-based Media

Lamont v. Postmaster General, a 1965 Supreme Court case, addressed whether an American citizen’s First Amendment rights were violated when the U.S. Post Office detained his copy of the Chinese news magazine Peking Review in 1963. In Lamont, the Supreme Court ruled that controlling the free flow of mail to the American public violates the First Amendment, just as controlling the free flow of ideas does.60 60. 381 U.S. 301, 306 (1965) [“just as the licensing or taxing authorities in the Lovell, Thomas, and Murdock cases sought to control the flow of ideas to the public, so here federal agencies regulate the flow of mail”]; See also Martin v. City of Struthers, 319 U.S. 141 (1943). The law that was struck down required recipients of “communist propaganda” to take certain steps to receive it. This was an impediment (as opposed to an obstacle) that the Supreme Court interpreted as a violation of the First Amendment. Although Lamont is considered a case about the rights of American listeners, it could be framed as addressing the regulation of American communicative infrastructure, specifically, the U.S. Postal Service.61 61. See, e.g., Wang, supra note 1; Grimmelmann, supra note 23 at 399–400.

Recognizing listeners’ rights was pathbreaking at the time and remains a relatively underdeveloped thread of First Amendment jurisprudence and scholarship.62 62. Wang, supra note 1. The ideal progression from Lamont would have been if First Amendment jurisprudence discussed Americans’ right to access foreign radio and television. In the period before Lamont and the world wars, foreign-owned radio was restricted in the United States.63 63. Sitaraman, supra note 5 at 1115–6. The concern at the time, although it included foreign propaganda, was primarily that the spectrum was limited and that foreign entities might find a way to run signal interference with American military communication.64 64. Adeno Addis, Who’s Afraid of Foreigners? The Restrictions on Alien Ownership of Electronic Media, 32 Colum. Hum. Rts. L. Rev 133, 147. The scarcity doctrine generally allowed for regulating broadcasting infrastructure in ways that were directly linked to the technological constraints of the limited spectrum.65 65. Turner Broadcasting System, Inc. et al. v. FCC I, 512 U.S. 622 (Supreme Court of the United States 1994). Legal norms regulating foreign ownership of broadcasting originated in these technological constraints and not in what might have been a helpful discussion of how threats to national security should be measured against Americans’ right to access foreign infrastructure.

TikTok Inc. v. Garland illustrates why thinking in terms of communicative infrastructure matters. TikTok Inc. v. Garland was about Americans’ access to foreign communicative infrastructure, even though it was never framed in this way for the court. Access to infrastructure affects access to ideas, and offers a sharper framing of what the justices might have been alluding to when they acted to protect the free flow of mail in Lamont.66 66. Lamont v. Postmaster General, 381 U.S. 301, at 306. Freedom of speech depends on the technological infrastructure that enables widespread democratic participation and supports the system of free expression.67 67. Jack M. Balkin, Digital Speech (2004) supra note 18 at 6. Social media audiences participate even through what looks like passive consumption. For example, watching a video about climate change is engagement because when people watch these videos, they tell the algorithm about their own likelihood, as well as their demographic group’s likelihood, of watching other videos like it. On TikTok, Lamont’s passive listener becomes a participant.

Modern communicative infrastructure is complicated because, unlike the postal service, it is privately owned. Nevertheless, protecting communicative infrastructure, including online platforms, is necessary to protect First Amendment values. The First Amendment cannot protect free speech values without the institutions, infrastructure, and industries that promote these values.68 68. Jack M. Balkin, Free Speech (2023) supra note 43. First Amendment law protects some kinds of communicative infrastructure, like the press and broadcast media, but fails to account for forms of communicative infrastructure that technology and online platforms create, although social media platforms are the dominant online communication model.

VII. Restricting TikTok 

The events leading to the TikTok restriction may be familiar to many readers. President Trump initiated the TikTok ban during his first term, and President Biden’s administration sealed it in place with the Protecting Americans from Foreign Adversary Controlled Applications Act (PAFACA).69 69. See Anupam Chander, Trump v. TikTok, 55 Vanderbilt Journal of Transnational Law 1145, 1148–56 (2022); Anupam Chander & Paul Schwartz, The President’s Authority over Cross-Border Data Flows, 172 U. Pa. L. Rev. 1989, 1990–1 (2024). PAFACA targets TikTok and potentially any other “foreign adversary controlled applications” by permitting the executive to designate them for restriction.70 70. 15 U.S.C. § 9901. Both administrations treated TikTok as presenting national security concerns because it was Chinese owned and controlled and therefore subject to the Chinese government, which they feared could influence data collection and algorithmic content targeting.

The D.C. District Court, before which the statute was challenged, deferred to the legislature’s and executive’s characterization of the problem, which can be summarized as follows: (1) China is a serious cyber espionage threat to the United States; (2) China has made significant efforts to gather data about Americans for intelligence purposes; (3) China has gathered data from Chinese companies as well as foreign companies in which it makes investments or from which it purchases data sets; (4) Chinese laws require Chinese companies to provide China with full access to their data for criminal and security investigations; and (5) Chinese companies are not independent in the way in which American companies are.71 71. TikTok v. Garland, 122 F.4th 930, 953–4 (2024).Additionally, the assistant director of the FBI asserted that (6) China tries to pre-position companies in the US that it can co-opt later and (7) engages in influence campaigns to counter and suppress opinions it objects to about China.72 72. Id. Additionally, the US government concluded, based on TikTok Global’s responsiveness to Chinese demands to censor content, that (8) TikTok would comply if the Chinese government asked it to manipulate content in the United States, especially since China maintains “a powerful Chinese Communist Party committee embedded in ByteDance,” (internal quotation marks omitted) and TikTok is reliant on and integrated with ByteDance.73 73. Id. at 954–5. Up to the point of the district court hearings, content-related concerns featured prominently among the justifications for banning TikTok.

In contrast, the Supreme Court focused on TikTok’s data collection, eliding the concerns about content and ignoring PAFACA’s regulation of expressive activity. Focusing exclusively on its data collection goal, the court argued that Congress would have passed the law even if it were not concerned about manipulative content.74 74. TikTok Inc. v. Garland, 604 U.S. __; Genevieve Lakier, The TikTok Ban and the Limits of the First Amendment, LPE Project (Jan. 22, 2025), https://lpeproject.org/blog/the-tiktok-ban-and-the-limits-of-the-first-amendment/. After framing data collection as the law’s goal, the Supreme Court argued that PAFACA regulated conduct rather than speech and applied the intermediate scrutiny test, not the strict scrutiny test.75 75. TikTok Inc. v. Garland, 604 U.S. __. The Court concluded that PAFACA, as applied to TikTok, satisfied the intermediate First Amendment scrutiny test because its objective—which the justices saw as preventing a foreign adversary, China, from leveraging its control of ByteDance to capture TikTok users’ personal data—qualified as an important government interest.76 76. Id. The justices also decided that PAFACA did not burden substantially more speech than necessary.77 77. Id.

VIII. Why TikTok Inc. v. Garland is Troubling

The Supreme Court’s decision should concern everyone interested in speakers’ rights, listeners’ rights, healthy public discourse, and the rights of freedom of thought and association. The justices’ conception of social media platforms led them to underestimate the potential impact of divestment on First Amendment interests. The Supreme Court does not appear to have carefully considered how to protect expressive or associational activity in TikTok Inc. v. Garland: It treated TikTok as static, ignoring its role as an evolving, ubiquitous infrastructure. In addition to concluding that the statute did not burden substantially more speech than necessary, the Supreme Court argued that the statute’s “TikTok-specific provisions” requiring divestment of the company are facially content neutral, in the context of whether strict scrutiny applies.78 78. Id. If the question before the justices had been framed as one of access to foreign infrastructure, different considerations might have been weighed against the national security interests animating PAFACA.

Major social media platforms, including TikTok, exhibit infrastructural features: they can be consumed non-rivalrously, enable downstream productive activity, and act as input into a wide range of goods and services.79 79. See Frischmann, supra note 3 at 61. Their ownership plays a role in how they do these things. It is possible that the leasing of the TikTok algorithm from Douyin will ensure that TikTok’s unique infrastructural role is not lost. However, the Supreme Court did not anticipate or require this outcome. In other words, it did not consider how Americans’ First Amendment interests are served by access to foreign communicative infrastructure.

The Supreme Court’s assessment of whether speech was affected and to what degree was based on an outdated understanding of how web-based platforms and Chinese platforms affect Americans’ First Amendment interests. Communicative platforms often offer entirely new ways of communicating. Social media platforms shape how communication is “transmitted, received, and recirculated,” and the “experiences and perceptions of what is ordinary and natural in the linked realms of private and public interconnection and sensemaking.”80 80. Cohen (2023), supra note 11 at 16. They also shape which people engage with each other and how people interact with bots. By affecting online association, they influence communication.81 81. For a discussion of the relationship between freedom of speech and freedom of association, see generally Ashutosh Bhagawat, Associational Speech, 120 Yale L.J. 978 (2011).

The infrastructural properties of each platform are distinct. Some platforms privilege content that elicits reactions from others, and other platforms (e.g., Bluesky) avoid rewarding reactions, choosing to treat the sharing of someone else’s post in exactly the same way as annotated sharing of the same post.82 82. See Wes Davis, Here’s some cool stuff you can do with Bluesky, The Verge, Nov. 16, 2024, https://www.theverge.com/24295933/bluesky-social-network-custom-how-to (about BlueSky’s simple chronological feed). Platforms also construct associations differently (although their choices may affect those of their competitors); TikTok’s method of profiling users has influenced Meta.83 83. McMahon, supra note 36. Safeguarding infrastructural diversity does not imply that platforms cannot be restricted or regulated, but shows that their differences play an important First Amendment role that should be considered while regulating them.

IX. Conclusion

The value of Americans’ right to access communicative infrastructure, including foreign communicative infrastructure, is under-considered in legal debate about communicative apps. Newcomers to the social media and artificial intelligence industries struggle to break through and are almost always purchased by the dominant firms, as WhatsApp and Instagram were.84 84. The AI industry also has a higher entry barrier because of the data and computational power required to train models. See Jennifer Cobbe, Michael Veale & Jatinder Singh, Understanding Accountability in Algorithmic Supply Chains1186 (2023), https://doi.org/10.1145/3593013.3594073. This leaves online communicative infrastructure very vulnerable to capture by private power.85 85. Tim Wu, The Master Switch 310 (2010). Focusing on citizens’ access to multiple sources of information and modes of engagement means that foreign communicative apps will remain an important source of alternative information, engagement, and competition. The judiciary’s willingness to defer to the legislative and executive branches’ characterization of communicative infrastructures as national security threats without careful consideration of their democratic value substantially undermines the First Amendment because of the scale of the speech and association affected.

If foreign-owned apps are easily restricted, communicative infrastructure will be left largely in the hands of a few American companies. Although these companies are not controlled by foreign states, they do have an interest in foreign markets and competing for profits. There is no reason to assume that they will never present or enable a national security threat. Using foreign ownership or control as a proxy for determining such a threat avoids the question of how communicative infrastructure should be regulated to mitigate national security threats and promote First Amendment values. Scholars have already proposed strategies, including fiduciary responsibility, interoperability, and anti-trust regulation, that might mitigate concerns about social media infrastructure.86 86. Cohen, supra note 11 at 16, 19. Layering additional regulations to address national security concerns is a more effective way to address concerns about foreign states than barring foreign-owned or foreign-controlled social media. Until American structural regulation improves, it is likely that apps capable of competing with dominant US communicative apps will only grow significantly in foreign countries with protected markets

Because non-US apps are shaped by foreign political, cultural, and economic forces, they create an environment for online engagement that American companies cannot create. They offer Americans what American companies have offered other publics in other countries for years: new ways to engage with ideas, with each other, and with people across borders. As communicative infrastructure, these apps are significant. To compel them to divest is to remove the quality that makes them unique: that their design reflects a distinct foreign influence.

Acknowledgments

I thank Maria Angel Arango, Jack Balkin, Katherine G. Bass, Liang Chen, Julie Cohen, Yuval Erez, Akriti Gaur, Jennifer Heinrichson, Margaret Jameel Jaffer, Hu, Shira Minsk, Michael McGovern, John Langford, Anat Leshnick, Lyrissa Lydsky, Yuping Lin, Dr. Margaret A. Oppenheimer and Cambridge Proofreading, John Peters, Robert Post, Alan Rozenshtein, Nishant Shah, David Schulz, Kalindi Vora, George Wang, Caleb Yong and Ethan Zuckerman. All mistakes are mine.

© 2026, Chinmayi Arun.

Cite as: Chinmayi Arun, The Right to Access Foreign Communicative Infrastructure, 26-4 Knight First Amend. Inst. (May 1, 2026), https://knightcolumbia.org/content/the-right-to-access-foreign-communicative-infrastructure [https://perma.cc/TQ6Q-48TB]. 

“There’s increasing awareness of the crucial role lawyers play during periods of democratic backsliding, in the United States and elsewhere, but not nearly enough public examination of it,” said Katy Glenn Bass, research director at the Knight First Amendment Institute. “We created this podcast to open that space, surface questions for further scholarship, and trace what responsibility looks like for lawyers in this moment.”

Authoritarianism is often understood as lawless, marked by constitutional rupture or institutional breakdown. But many of the most effective assaults on democracy operate through legal systems themselves. Around the world, leaders have used laws, courts, and institutional processes to consolidate power, restrict dissent, and weaken accountability while maintaining the appearance of legality. Similar pressures are emerging in the United States, raising urgent questions about how the legal profession responds when these dynamics take hold closer to home.

“One of the central concerns that motivated this project is that we’ve spent a great deal of time studying courts and political institutions, but far less time examining lawyers themselves,” said Madhav Khosla, senior fellow at the Knight First Amendment Institute and professor at Columbia Law School. “Lawyers are often the first to see how legal systems are being used or misused, yet we haven’t fully reckoned with the role they play in shaping those outcomes.”

The podcast is part of a broader collaboration between the Knight Institute and Khosla examining how the legal profession is evolving in response to democratic backsliding. The project includes a scholarly essay series and public convenings exploring how professional norms, ethical frameworks, and institutions such as bar associations respond to political pressure.

The first episode, “What Does Legal Authoritarianism Look Like?,” features Kim Lane Scheppele, the Laurance S. Rockefeller Professor of Sociology and International Affairs at the Princeton School of Public and International Affairs. The episode explores how authoritarianism can take hold through ordinary legal processes and what it means for lawyers when legal systems depart from their constitutional commitments.

Future episodes will feature leading scholars and practitioners, including Lawrence Lessig, Deborah Pearlstein, and David Dyzenhaus examining the role of the legal profession across a range of political and institutional contexts.

“Conversations about democratic backsliding and the role of law can feel abstract, even as these pressures are playing out in real time,” said Candace White, deputy director of communications and digital strategy at the Knight First Amendment Institute and executive producer of “Lawyering Without Law.” “This series makes those pressures visible, helping people understand how legal systems are being used and what’s at stake.”

“Lawyering Without Law” is available on Apple Podcasts, Spotify, and other platforms, with new episodes released every other Friday. The podcast is hosted by Katy Glenn Bass and Madhav Khosla and produced and engineered by Dustin Foote, with Candace White serving as executive producer. Fact-checking is by Connor Menzies and Sophia Rojas; music is from Envato Elements, and artwork is by Jay Vollmar.

Listen to Episode 1 here.

Learn more about the series here.

Read more about Khosla’s research project here.

The Knight Institute has also produced four award-winning podcasts: “What Happens When Social Media Collides with the First Amendment?,” “War & Speech,” “Speech & the Border,” and “The Bully’s Pulpit: Trump v. the First Amendment.” Explore them here.

For more information, contact: Adriana Lamirande, adriana.lamirande@knightcolumbia.org 

The following can be attributed to Jameel Jaffer, executive director at the Knight First Amendment Institute at Columbia University:

“The FCC has no authority to cancel broadcasters’ licenses because of their perceived political views. But this isn’t just about the rights of Disney and ABC. President Trump is trying to consolidate control over what Americans see and hear on the radio, television, and social media. If he gets his way, we’ll have only government-aligned media organizations that broadcast only government-approved news and commentary. It would be difficult to imagine an outcome more corrosive to democracy or more offensive to the First Amendment.”

For more information, contact: Adriana Lamirande, adriana.lamirande@knightcolumbia.org 

The following can be attributed to Xiangnong (George) Wang, staff attorney at the Knight First Amendment Institute at Columbia University:

“We welcome the Mayor’s veto of legislation that would have charged the NYPD with setting the rules around lawful political protest near thousands of schools and educational facilities across the city. This measure risks chilling and even criminalizing a wide range of speech the First Amendment protects, particularly at some of the city’s most vital sites for public discourse. Governments shouldn’t entrust law enforcement with determining the limits of political expression.”

For more information, contact: Adriana Lamirande, adriana.lamirande@knightcolumbia.org

The following can be attributed to Carrie DeCell, senior staff attorney and legislative advisor at the Knight First Amendment Institute at Columbia University: 

“The State Department is excluding tech researchers from the United States because of their constitutionally protected work, and this newly disclosed memo only underscores the unbounded scope and unconstitutionality of the policy. That the Trump administration claims its censorial policy is intended to combat censorship is the height of doublespeak.”

In March, the Knight Institute and Protect Democracy filed a lawsuit on behalf of the Coalition for Independent Technology Research (CITR) challenging the constitutionality of the policy. The complaint argues that the policy violates the First Amendment and chills independent research and reporting about social media and other internet platforms.

Read more about the case here.

For more information, contact: Adriana Lamirande, adriana.lamirande@knightcolumbia.org 

Spyware tools like Pegasus, developed by NSO Group, enable covert access to journalists’ phones, communications, and personal data, often through “zero-click” attacks that require no user interaction. Farid Johnson emphasized that these technologies endanger journalists and their sources while disrupting newsroom operations, chilling investigative reporting, and limiting the public’s access to information.

Farid Johnson also highlighted the legal barriers spyware victims face in U.S. courts, noting that procedural hurdles have kept many cases from reaching the merits. She urged Congress to amend the Computer Fraud and Abuse Act to clarify that U.S. courts should hear claims involving spyware attacks that exploit U.S.-based technology infrastructure, ensuring a meaningful pathway to redress.

The hearing also featured testimony from Sergio Arauz, deputy editor-in-chief of El Faro and a plaintiff in Dada v. NSO Group—the Institute’s lawsuit on behalf of journalists targeted with Pegasus spyware. He described escalating repression under El Salvador’s ongoing state of exception, including surveillance, exile, and the criminalization of journalism. His testimony underscored the real-world consequences of the dynamics Farid Johnson identified, linking the use of spyware to broader patterns of democratic erosion and the silencing of dissent.

Read Farid Johnson’s full statement submitted for the record, here. 

Read Sergio Arauz’s full testimony, here. 

Watch a recording of the full hearing below.

Around the world, leaders like Recep Tayyip Erdoğan in Turkey and Viktor Orbán, the former prime minister of Hungary, have used legal systems, rules of law, and institutional practices to consolidate power, restrict dissent, and hollow out democratic accountability from within. That pattern is becoming more visible in the United States, where mounting political pressure on courts, lawyers, and legal institutions is raising urgent questions about the role of the legal profession in moments of democratic crisis. 

“Lawyering Without Law,” a bi-weekly podcast from the Knight Institute, interrogates the unique and important role that lawyers play in defending democracy, or in facilitating the slide into authoritarianism. Hosted by Knight Institute Senior Fellow and Columbia Law Professor Madhav Khosla and the Knight Institute’s Research Director Katy Glenn Bass, the series brings together scholars, litigators, and practitioners to explore these dynamics across historical and contemporary contexts. Drawing on global examples of democratic backsliding, each episode connects these developments to the United States and outlines what is at stake for the legal profession and for democracy itself. 

Read more about Khosla’s research project with the Knight Institute examining the crucial role that lawyers can play in preserving democratic freedoms and institutions here.

“Lawyering Without Law” is available on Apple, Spotify, and wherever you get podcasts.

Listen, subscribe, and leave a review. We’d love to know what you think.

Subscribe

Episode One: What Does Legal Authoritarianism Look Like?

What does authoritarianism look like when it operates through law? In the first episode of “Lawyer Without Law,” hosts Katy Glenn Bass and Madhav Khosla speak with Princeton University Professor Kim Lane Scheppele. They explore historic examples of the legal profession’s role in democratic backsliding around the world and in the United States. They examine how legal systems can consolidate power while maintaining the appearance of legitimacy—and what that means for lawyers and legal institutions as democratic norms come under strain.

Transcript: Ep. 1. 

Further Reading

• At Harvard, Mitt Romney Warns Against ‘Authoritarian’ Presidential Power, by Schuyler Velasco, Harvard Magazine (4/16/2026)
• What Are We Living Through in Trump 2.0? Here Are 3 Possibilities. by Jedediah Britton-Purdy, David Pozen, and John Guida, The New York Times (3/13/2026)
• Kim Lane Scheppele on Hungary, Paul Krugman (Substack) (4/18/2026)
• Collateral Knowledge: Legal Reasoning in the Global Financial Markets, by Fleur Johns and Annelise Riles, European Journal of International Law (12/17/2012) 
• “Ibram X. Kendi and Heather McGhee on the world’s most dangerous conspiracy theory,” @penguinrandomhouse (YouTube) (3/16/2026)
• The professions in America today, by Howard Earl Gardner and Lee S. Shulman, Dædalus (2005)

The following can be attributed to Xiangnong (George) Wang, staff attorney at the Knight First Amendment Institute at Columbia University:

“We’re disappointed that the City Council passed these bills, which would task the NYPD with writing the rules on where and how people may engage in lawful political protest near the city’s thousands of schools, educational facilities, and places of worship. These measures risk chilling and even criminalizing a wide range of speech protected by the First Amendment around some of the city’s most important sites for public discourse. Governments shouldn’t task law enforcement with determining the limits of political expression. We urge the Mayor to do everything in his power to ensure these bills do not become law.”

For more information, contact: Adriana Lamirande, adriana.lamirande@knightcolumbia.org 

The amicus brief warns that adopting Amazon’s interpretation of the laws at issue would chill journalism and research that serves the public interest. Journalists and researchers increasingly use automated digital tools—including scrapers, browser extensions, and other AI-powered tools—to study online platforms and the ways in which they shape public discourse. Like Perplexity’s AI agents, these tools often rely on users to voluntarily provide them with access to their accounts. The brief argues that the Computer Fraud and Abuse Act and California’s analog should not be interpreted to prohibit these user-directed activities, because doing so would implicate the basic tools of digital investigation necessary for research and journalism online. 

Status: Briefing ongoing. 

Case Information: Amazon.com Servs., LLC v. Perplexity AI, Inc., No. 26-1444 (9th Cir.)

For years, immigration judges regularly spoke at conferences, guest lectured at universities and law schools, participated in immigration-law trainings, and spoke to local community groups, all in their personal capacities. But starting in 2017, the Executive Office for Immigration Review issued a series of speaking-engagement policies that sharply curtailed their ability to speak publicly in their personal capacities

The lawsuit argues that the currently operative policy violates the First Amendment right of immigration judges to speak publicly on matters of public concern, and the First Amendment right of the public to hear them. It also argues that the policy is void for vagueness under the First and Fifth Amendments.

On June 3, 2025, the Fourth Circuit vacated the district court’s decision dismissing the case for lack of subject matter jurisdiction, and remanded for further proceedings consistent with its opinion.

Status: Briefing complete on the government’s petition for certiorari and NAIJ’s cross-petition for a writ of certiorari.

Case information: Nat'l Ass'n of Immigration Judges v. Owen, No. 1:20-cv-00731 (E.D. Va.), No. 20-1868 and 23-2235 (4th Cir.), Margolin v. Nat'l Ass'n of Immigration Judges, No. 25A662, No. 25-767, 25-1009.

The following can be attributed to Jameel Jaffer, executive director at the Knight First Amendment Institute at Columbia University:

“News organizations have a First Amendment right to publish stories about matters of public importance—including stories the government would prefer to suppress. President Trump’s threat to force journalists to disclose their sources raises serious press freedom concerns because journalists’ ability to do their work turns in part on their ability to protect their sources’ identities. President Trump's threat should be understood as an effort to intimidate the press and to prevent journalists from doing work the public needs them to do.”

For more information, contact: Adriana Lamirande, adriana.lamirande@knightcolumbia.org 

Throughout the Institute’s half-day program—“Can Middleware Save Social Media?”—those questions played out across three panel discussions about online speech and the rules governing social media platforms.

As Jameel Jaffer, executive director of the Knight Institute, said at the outset: The question is not just whether social media can be saved, but what we are saving it from—and who gets to do the saving.

Defining Middleware

Panelists described middleware not as a single product, but as a layer of tools that sits between users and platforms, shaping how information is filtered and delivered. Early examples such as ad blockers showed how users could modify the content that platforms deliver. Newer systems go further, allowing users to curate their own feeds or apply alternative algorithms.

However, that shift comes with constraints, including legal barriers that limit interoperability, privacy risks, and the reality that platforms still hold enormous structural power and control the terms of engagement. Middleware, in this sense, operates within and around existing systems rather than replacing them.

In conversation with the Institute’s Ramya Krishnan, Daphne Keller of Stanford University and Richard Reisman of the Foundation for American Innovation explored these dynamics, emphasizing how middleware can redistribute control to users without inviting direct government intervention. At the same time, they underscored that middleware’s effectiveness depends on the surrounding legal and technical environment—whether platforms are required, or even willing, to allow interoperability. Without that, the promise of user control remains contingent, not guaranteed.

Middleware’s Promise and Its Limits

The keynote conversation, moderated by the Institute’s Policy Director Nadine Farid Johnson, brought together Olivier Sylvain, senior policy research fellow at the Knight Institute and professor of law at Fordham University, and Ethan Zuckerman of the University of Massachusetts Amherst to examine what middleware can actually do.

For Zuckerman, middleware offers a tangible but limited path forward. “The problem with middleware is that it is a partial, incomplete, imperfect solution,” he said. Still, he described it as “a way of showing that another world is possible,” one that could help build the case for broader reform.

At one point, the conversation turned to whether users can realistically take control of their own online experience. Zuckerman mentioned tools that prompt reflection—like weekly screen time reports on smartphones—as evidence that users might engage more intentionally with their media environments if given the opportunity.

Sylvain challenged that premise, shifting the focus from individual behavior to system design.

“There is an underlying pathology, a structural problem that middleware does not answer,” Sylvain said, arguing that the current system is defined by information asymmetries—what platforms know, what users do not, and how that gap is leveraged to shape how content is delivered and consumed. He warned that focusing too squarely “on the user as a solution … doubles down on the problem we have now.”

The exchange crystallized a central tension: whether meaningful reform begins with empowering users to reshape their own experience, or with restructuring the systems that shape it for them.

Considerations for Policymakers

In conversation with the Institute’s Ryan Morgan, panelists Renée DiResta of the Stanford Internet Observatory, Anna Lenhart of Common Sense Media, and Luke Hogg of the Foundation for American Innovation examined the policy landscape surrounding middleware and platform governance, with a focus on what interventions could meaningfully shift power in the digital public sphere.

They discussed a range of legislative proposals at the federal and state levels that could facilitate middleware development and adoption, alongside broader efforts to strengthen data privacy protections and address the structural dynamics that shape how information is distributed and consumed online.

The discussion made it clear that middleware alone cannot resolve the deeper forces shaping online discourse.

What middleware can do, however, is expose the limits of the current system and sharpen the case for structural reform. The future of online discourse will not be decided solely by platforms or policymakers, but by whether new systems can redistribute control in ways that are both meaningful and durable.

The larger question is whether the digital public sphere can be restructured so that the power over what we see is not concentrated by a handful of platforms, but more broadly shared. Middleware offers one entry point.

The following can be attributed to Nadine Farid Johnson, policy director at the Knight First Amendment Institute at Columbia University:

“ICE’s use of powerful spyware tools raises serious civil liberties concerns. Spyware enables covert and often unlimited access to smartphone data, posing significant risks to free speech and privacy. Experience shows these tools are highly susceptible to misuse; they’ve already been used to target journalists, human rights advocates, and political dissidents around the world. Democracies should not be in the business of deploying spyware against their populations. We appreciate Representative Lee’s focus on this issue and again urge Congress to step in to limit the circumstances in which spyware technology can be used.”

In 2022, the Knight Institute filed a lawsuit on behalf of journalists and other members of El Faro—one of Central America’s foremost independent news organizations, based in El Salvador—who were targeted with spyware attacks using NSO Group’s Pegasus technology. In July of last year, the U.S. Court of Appeals for the Ninth Circuit held that the U.S. District Court for the District of Columbia had abused its discretion in dismissing the lawsuit and remanded the case for further consideration. Read more about that case, Dada v. NSO Group, here.  

For more information, contact: Adriana Lamirande, adriana.lamirande@knightcolumbia.org 

While Section 230 shouldn’t be considered sacrosanct, its repeal would do little to address the problems lawmakers are trying to solve. In some ways, it would make the problems worse. What’s needed instead is a new legislative approach grounded in structural reform—one that protects users’ privacy, allows users to engage with platforms on their own terms or leave them more easily, and makes platforms more transparent and accountable. 

Lawmakers, parents and advocates have raised serious concerns about harmful content, particularly as it affects minors, as well as the power of large platforms and the inability to hold them accountable for their decisions. Congress is now considering bills that would sunset Section 230. The Senate Committee on Commerce, Science, and Transportation recently held a hearing, at which I testified, on how to move forward.  

Framing the discussion about Section 230 as a choice between keeping the provision intact or scrapping it altogether misses a key point: Repeal wouldn’t fix many of the concerns raised. 

Here’s why: Many of the harms driving calls for reform are tied to speech the First Amendment already protects. That constraint is not incidental. It shapes what Congress can do. As the Supreme Court recently reaffirmed, platforms’ editorial decisions about whether and how to display content are protected by the Constitution. 

Repealing Section 230 would not change the fact that much of the content lawmakers are concerned about—often described as “lawful but awful” speech—would remain protected. The government cannot prohibit that speech, nor can it compel platforms to do so. 

What repeal would do is change the incentives that shape platform behavior, to the detriment of users and public discourse. 

Without Section 230, platforms would face increased legal risk for hosting user speech, including defamatory claims alleging wrongdoing by identifiable individuals. The First Amendment doesn’t protect defamation. Truth is a defense to liability, but platforms cannot reliably determine the truth of such claims at scale. They would therefore have strong incentives to remove any claims that might give rise to a defamation lawsuit. The result wouldn’t be a safer or meaningfully improved online environment, but one in which lawful, often socially valuable speech is taken down more frequently. 

The effects wouldn’t be evenly distributed across platforms. While the largest platforms may be able to absorb the costs of increased liability, smaller or newer platforms may not. Community-driven sites and emerging services would be particularly vulnerable. The likely outcome would be an even more concentrated online landscape, with fewer options for users and less competition. 

None of this is to defend the status quo. Few would argue that the digital public sphere is working for Americans or for our democracy. The question is how to respond to a rapidly evolving technology landscape in ways that are both effective and consistent with the First Amendment. 

The most productive path forward lies in structural reform that targets the underlying features of the current system that contribute to online harms without creating incentives for platforms to remove lawful speech or giving the largest platforms even more control over online discourse. 

Lawmakers could require greater transparency about how platforms operate, including how they collect and use consumer data and how their systems shape what users see. Congress could also establish protections for journalists and researchers who study platforms in the public interest, such as those outlined by my organization in the Knight Institute’s safe harbor proposal. 

Platforms are successful in maintaining user engagement in part by relying on the extensive information they gather. Lawmakers could strengthen privacy protections by limiting what information platforms collect about users and how that information is shared, including by restricting the sale of user data. They could also give users more control over their online experience, including by making it easier for them to move their data and connections across platforms or interact with users of competing services. 

Congress could enact these reforms independently of Section 230 or condition its protections on compliance with these requirements. Either way, the platforms would have little choice but to respect the privacy of their users, provide greater transparency into how they operate and give users greater control over their online lives—all while preserving the space for public discourse. 

Whether to repeal Section 230 is not the right question. The more important question is how to address online harms without undermining free expression.  

The better course is to pursue targeted reforms that address real concerns about the online experience while respecting the constitutional limits that govern speech in the U.S. 

Repealing Section 230 will not achieve that. Structural reform can. 

The lawsuit alleges that the policy violates the First Amendment by penalizing particular viewpoints and deterring independent research about social media and other internet platforms. It also raises claims under the Fifth Amendment and the Administrative Procedure Act.

CITR’s motion asks the court to halt enforcement of the policy while the case proceeds, explaining that its members are already self-censoring by curtailing research, avoiding speaking publicly about their work, and limiting their participation in advocacy efforts for fear of being targeted by the government for their public-interest work.

CITR’s members include research institutions, academics, and journalists who study digital platforms and their societal impacts. Their work seeks to inform public debate so that consumers, advertisers, platforms, and policymakers can make informed decisions about emerging technologies.

The following can be attributed to Carrie DeCell, senior staff attorney at the Knight First Amendment Institute:

“The Trump administration claims that its new exclusion and deportation policy counters censorship, but it is itself censorship. In targeting independent researchers for studying and reporting on social media and other internet platforms, the policy punishes work that the First Amendment protects—and work that the public needs to understand how the platforms are shaping our society.”

The following can be attributed to Clare Melford, co-founder of the Global Disinformation Index, a CITR member organization:

“Because of the policy, I’ve been prevented from traveling to the United States. I had to cancel meetings with colleagues and funders and postpone work that depends on in-person collaboration. That kind of disruption slows research, breaks down partnerships, and limits the exchange of ideas across borders.”

The following can be attributed to Brandi Geurkink, executive director of the Coalition for Independent Technology Research:

“Because of the government’s censorship policy, researchers are pulling back on studying critical topics and avoiding speaking publicly about their work, because they fear  they could be  detained or deported because of what they say. If this assault on research continues, people will be left without independent information about the impacts of AI and other digital platforms on our societies—at precisely the moment when we need it most.”

Read the preliminary injunction motion here.

Read more about the case here.

Lawyers on the case include Carrie DeCell, Raya Koreh, Kiran Wattamwar, Anna Diakun, Katie Fallow, Alex Abdo, and Jameel Jaffer, for the Knight First Amendment Institute, and Naomi Gilens, Nicole Schneidman, Scott Shuchart, and Deana El-Mallawany, for Protect Democracy.

For more informtaion, contact: Adriana Lamirande, adriana.lamirande@knightcolumbia.org. 

In seeking avenues to address these concerns without impinging on First Amendment rights, many free speech and technology advocates are looking to middleware, third-party software that operates between users and platforms. Middleware’s proponents say that this kind of software could permit users to enhance their ability to shape their online experience, including by curating their timelines and by exercising more control over what data they share. Skeptics say that middleware raises privacy concerns of its own and that more fundamental changes are needed to address the pathologies of the digital public sphere. Can middleware really deliver the improvements that its boosters envision? 

On March 27, 2026, the Knight Institute will host “Can Middleware Save Social Media?” a half-day convening focusing on these questions and the future of these go-between tools. This convening is a collaboration between the Knight Institute and the Institute’s Senior Policy Fellow Olivier Sylvain and will include discussions with academics, policymakers, and technologists. 

Registration is required to attend in-person or to watch the livestream. 

The following can be attributed to Nadine Farid Johnson, policy director at the Knight First Amendment Institute at Columbia University:

“Tasking the NYPD to write the rules on where and how people may engage in lawful political protest risks chilling and criminalizing a wide range of activities protected by the First Amendment. It’s especially alarming because these rules would cover schools, other educational facilities, and places of religious worship, which are the sites of some of the city's most vital public discourse.”

For more information, contact: Adriana Lamirande, adriana.lamirande@knightcolumbia.org 

All this has given rise to a formally complicated, almost flow chart-like model for evaluating First Amendment challenges. Does the law regulate speech, or merely conduct? Is it content-based? If so, can the government satisfy strict scrutiny? And if it’s content-neutral, will the parties merely just fight over tailoring? Because of the judicial skepticism of laws that regulate speech, particularly content-based regulations, challengers to government action fight strenuously to persuade the court to identify the action as speech. The government, for its part, will almost always begin with a claim that there’s nothing to see here: this law merely regulates conduct or is government speech, and the court should let it stand.

Formalism may have its virtues in free speech cases, particularly if a rules-based framework minimizes chilling effects and allows speakers to express themselves with less fear of reprisal. But the current overreliance on rules, in my view, has created a predictable result—parties hostile to government regulation try to jam their claims into the First Amendment, hopeful that they can pull off a maneuver that invalidates the entire scheme as content-based, or perhaps not even properly tailored under intermediate scrutiny.

This is where we have arrived: a system in which any legislator or regulator thinking of drafting a a law that regulates information or data has to think “Is this going to survive the Supreme Court’s view of the First Amendment?” And absent a radical narrowing of what falls into First Amendment-protected speech—a move the Court could make, but seems unlikely—I see only one manageable, principled path forward to allow for government regulation of the large swaths of the economy that supposedly implicate speech. The Court must embrace free speech balancing tests.

Calling for balancing inquiries in speech cases may seem politically impossible, absurd, or ill-advised. The need for rules to provide clarity and predictability for speakers, and to limit the possibility of judges upholding the censorship of speech they dislike, has become a near orthodoxy in free speech cases.6 6. See United States v. Stevens, 559 U.S. 460, 470 (2010) (“The First Amendment's guarantee of free speech does not extend only to categories of speech that survive an ad hoc balancing of relative social costs and benefits. The First Amendment itself reflects a judgment by the American people that the benefits of its restrictions on the Government outweigh the costs. Our Constitution forecloses any attempt to revise that judgment simply on the basis that some speech is not worth it.”). Some doctrinal frameworks do incorporate balancing, as in the test that applies to public employees speaking on matters of public concern. Pickering v. Board of Education, 391 U.S. 563 (1968). In my view, though, the Court has itself abandoned this commitment to rules even as First Amendment doctrine has become more convoluted, contradictory, and opaque. Whatever virtues clear rules provided, they have faded.

Consider two decisions from the Court’s last term. Free Speech Coalition v. Paxton concerned a Texas law mandating age verification for access to explicit content online.7 7. 606 U.S. 461 (2025). Justice Thomas’ majority distinguishes the case from ACLU v. Ashcroft,8 8. 542 U.S. 656 (2004). a 2004 case striking down a similar federal law, by arguing that a burden on adults to access explicit content differs from a ban; no matter that the Court had already said that burdens do not qualify the level of scrutiny in U.S. v. Playboy Entertainment Group.9 9. See United States v. Playboy Entertainment Group, 529 U.S. 803, 826 (2000) (“[S]pecial consideration or latitude is not accorded to the Government merely because the law can somehow be described as a burden rather than outright suppression.”) The rule of Ashcroft, Playboy Entertainment Group, and other cases became something that the court circumvented because it could.

In TikTok v. Garland,10 10. 606 U.S. 56 (2025). the Court evaluated a federal law that effectively forced TikTok to either sever ties with its Chinese holding company or face a shutdown within the United States. The social media company argued that the law violated the First Amendment; the federal government countered that it had vital interests in preventing foreign governments from manipulating Americans via social media feeds and in collecting data on American users. Justice Kagan noted at oral argument that the former sounded like a content-based restriction; the latter, however, carried the day as the justification for upholding the law. But by smuggling a mixed-motives justification rule into the First Amendment—in which a law with two motives, one that violates the First Amendment and one that does not, can still survive—the Court created a world in which courts can pick and choose which motives to investigate and which to ignore.

Beyond these two cases, other recent decisions show how the Court has allowed its supposedly firm doctrinal rules to permit discretionary choices by judges. NIFLA v. Becerra11 11. 585 U.S. 755 (2018). weakened the standard from Zauderer v. Office of Disciplinary Counsel12 12. 471 U.S. 626 (1985). by allowing challengers to seek invalidation of transparency requirements on “controversial” topics, ignoring how challengers can actually manufacture controversy to begin with. Americans for Prosperity v. Bonta13 13. 594 U.S. 595 (2021). ignored the longstanding ban on “subjective ‘chill’” as a basis for standing in First Amendment cases. 303 Creative v. Elenis14 14. 600 U.S. 570 (2023).created an exemption for antidiscrimination laws applying to businesses engaged in “expressive activity,” conceding that determining what falls in that category might “raise difficult questions.”15 15. Id. at 599. If the Court once feared allowing standards and judicial discretion in First Amendment cases, it seems to recently have gained some courage—though it hasn’t actually acknowledged it.

Rather than stop this move into standards-based reasoning, I in fact think the Court should more explicitly adopt it. Justice Breyer, on multiple occasions, called for means-ends balancing in First Amendment cases to allow for consideration of governmental motives and speech interests, eschewing “a mechanical use of categories.”16 16. Reed v. Town of Gilbert, 576 U.S. 155, 179 (2015) (Breyer, J., concurring in the judgment). Breyer even managed to get the Court to adopt balancing tests in two of his last First Amendment majority opinions, Mahanoy Area School District v. B.L.17 17. 594 U.S. 180 (2021).and Shurtleff v. City of Boston.18 18. 596 U.S. 243 (2022). In Mahanoy, Breyer set forth a number of factors to determine whether a K-12 school could regulate off-campus speech of students.19 19. Mahanoy, 594 U.S. at 189 (“We can, however, mention three features of off-campus speech that often, even if not always, distinguish schools’ efforts to regulate that speech from their efforts to regulate on-campus speech. Those features diminish the strength of the unique educational characteristics that might call for special First Amendment leeway.”).And in Shurtleff, Breyer fashioned a contextual inquiry to determine whether the government speech doctrine applied.20 20. Shurtleff, 596 U.S. at 252 (“[W]e conduct a holistic inquiry designed to determine whether the government intends to speak for itself or to regulate private expression. Our review is not mechanical; it is driven by a case's context rather than the rote application of rigid factors.”). While these balancing tests don’t change the core of free speech doctrine, they show a perhaps surprising potential to incorporate standards and balancing into other First Amendment areas.

Beyond the need to better reflect some of the Court’s actual recent free speech decisions, incorporating standards furthers political goals that a reimagined First Amendment must take into account. A brittle, harsh First Amendment system of categories and stark rules makes it difficult, if not impossible to regulate in the areas most essential to promote a contemporary democratic society, including campaign finance, anti-discrimination, and information governance. The Court has recently, belatedly, and partially adopted a Breyerian fondness for standards and balancing in free speech cases. Directly acknowledging and continuing that shift would allow for a healthier First Amendment environment.

Beyond the courts, entities that use First Amendment principles to inform their own speech regulations (such as social media companies, private universities, and some private employers) should more explicitly and transparently adopt balancing frameworks, which could help socialize common practices in the private sphere. Legislators and regulators at all levels of government should contemplate and enact legislation that might force the Supreme Court to reconsider its excessively formalized doctrines. As many social movements have taught in other areas of rights and liberties, while the Court will probably not, on its own, develop a pro-democracy First Amendment, we can attempt to guide it to that necessary end.

The letter raises particular concern about Carr’s use of the FCC’s “public interest” standard as a tool to target viewpoints disfavored by the Trump administration. It explains that the standard does not authorize the FCC to police editorial decisions or penalize protected speech, and warns that vague and selective enforcement risks chilling lawful reporting. The First Amendment, the letter emphasizes, does not permit the government to coerce private actors or reshape the content of the news. The coalition calls on the FCC to withdraw its threats and make clear that it will not use its regulatory authority to influence or control press coverage.

In 2024, the Knight Institute launched “Jawboning and the First Amendment,” a research initiative examining how informal government pressure can function as a form of censorship, why it matters, and what legal and policy responses can address its harms.

Read today’s letter here.